The Hershey Company releases new Cookie Layer Crunch flavour

first_imgThe Hershey Company releases new Cookie Layer Crunch flavourPosted By: News Deskon: December 05, 2017In: Confectionery, Food, Industries, New products, SnacksPrintEmailThe Hershey Company has launched a triple chocolate variant of its Cookie Layer Crunch product, available in the US.Made from chocolate crème, chocolate cookie bits and milk chocolate, the product joins three other Cookie Layer Crunch flavours: caramel, vanilla crème, and mint crème.Initially launched in December 2016, the range features Hershey’s chocolate with layers of cookie pieces.The line is available in an assortment of sizes: 1.4oz bar (all flavours), 2.1oz king size bar (offered in triple chocolate, caramel, and mint), 3.5oz extra-large bar (caramel only), and the 6.3oz nine-piece bag (all flavours).Hershey’s Franchise senior brand manager Edward Edson said: “Following the successful launch of Hershey’s Cookie Layer Crunch last year, it was evident that Hershey fans couldn’t get enough of the delicious layers of our newest innovation in chocolate treats.“Due to the popularity of the product, the brand wanted to expand our flavour profile of Hershey’s Cookie Layer Crunch to offer more delicious combinations and from that we developed triple chocolate.”Alongside the new flavour offering, The Hershey Company is rolling out a new advertising campaign called ‘Get Layered’. It features commercials which are meant to glorify the importance of layers in the chocolate treat and in everyday life.Edson added: “The new campaign taps into the insight that consumers find adding layers to food/life makes the experience better. This idea sparked, what we believe to be, compelling and unique creative that will air across TV, social and online.”Share with your network: Tags: The Hershey CompanyUnited Stateslast_img read more

JBT Corporation buys tray sealing specialist Proseal in £220m deal

first_imgJBT Corporation buys tray sealing specialist Proseal in £220m dealPosted By: Contributoron: April 30, 2019In: Business, Food, Industries, Manufacturing, Mergers & Acquisitions, PackagingPrintEmailJBT Corporation has signed an agreement to acquire Proseal, a provider of tray sealing technology for the food industry, in a deal worth £220 million.Headquartered in Adlington, UK, Proseal has additional production facilities in Richmond, Virginia, and Melbourne, Australia.Proseal’s tray sealing packaging technology is used across a range of food products, including ready meals, fresh produce, sandwiches, and proteins, to enhance freshness by sealing the product with film over a preformed tray.JBT said Proseal has been at the “leading edge of developing environmentally friendly packaging that minimises the use of plastics while reducing food waste by extending the product’s shelf life”.Tom Giacomini, JBT CEO, said: “The acquisition of Proseal represents an important expansion of JBT’s capabilities, adding significantly to our end-of-line market position and advancing our strategy of providing full-line customer solutions. Moreover, we believe Proseal’s technology affords growth opportunities around the globe.”Carlos Fernandez, JBT executive vice president and president, liquid foods, added: “Tray sealing is relevant to many of our food-tech customers, particularly in the rapidly expanding market for convenience foods.”The acquisition is expected to boost JBT’s annual revenue by approximately $100 million. Closing of the deal is scheduled for the second quarter of 2019.Last July, JBT acquired Netherlands-based food equipment and solutions provider FTNON for €32 million to expand in the fruit and vegetable market.FTNON, which is headquartered in Almelo, serves the fresh-cut and thermal food market with a focus on peeling, steaming, de-coring, washing, slicing and blanching.Share with your network: Tags: JBT CorporationProsealUKUSlast_img read more

Nayax introduces loyalty service and refunds to its payment app

first_imgNayax introduces loyalty service and refunds to its payment appPosted By: Martin Whiteon: May 21, 2019In: Business, Industries, Refreshment, TechnologyPrintEmailNayax has updated its Monyx Wallet payment app to include a consumer loyalty programme and a refund service for unattended retail locations such as vending machines.Consumers can use the Nayax Wallet app by loading cash onto the app, and can then pay for products at Monyx-integrated machines using their smartphones.With the new update, Monyx Wallet 2.0 provides a range of loyalty benefits to consumers including discounts, cash back and bonus credit, and allows operators to run their own marketing campaigns and loyalty programmes for staff in closed-loop environments.According to Nayax, third-party loyalty apps can also benefit from Monyx Wallet’s payment and engagement features via an SDK framework, which integrates Nayax’s digital wallet capabilities into existing apps.Yair Nechmad, CEO of Nayax said: “Nayax believes relationships are key to retaining regular customers.“A boost in sales is beneficial for any business but repeat business is better for long term success.“For consumers at unattended machines the business owner is anonymous – Monyx Wallet’s marketing tools help operators develop more of a relationship with their consumers, and support operators in promoting their business.“By nurturing loyalty, operators set themselves up for a future with consistent revenue.”Share with your network: Tags: cashless paymentsNayaxlast_img read more

Kentucky teachers stage sick out to protest against political shadiness

first_imgKentucky teachers stage ‘sick out’ to protest against political ‘shadiness’ Share on Pinterest Shares108108 Since you’re here… Share on Messenger Support The Guardian news Associated Press Share on LinkedIn Share on Facebook … we have a small favour to ask. The Guardian will engage with the most critical issues of our time – from the escalating climate catastrophe to widespread inequality to the influence of big tech on our lives. At a time when factual information is a necessity, we believe that each of us, around the world, deserves access to accurate reporting with integrity at its heart.More people are reading and supporting The Guardian’s independent, investigative journalism than ever before. And unlike many news organisations, we have chosen an approach that allows us to keep our journalism accessible to all, regardless of where they live or what they can afford. But we need your ongoing support to keep working as we do.Our editorial independence means we set our own agenda and voice our own opinions. Guardian journalism is free from commercial and political bias and not influenced by billionaire owners or shareholders. This means we can give a voice to those less heard, explore where others turn away, and rigorously challenge those in power.We need your support to keep delivering quality journalism, to maintain our openness and to protect our precious independence. Every reader contribution, big or small, is so valuable. Support The Guardian from as little as $1 – and it only takes a minute. Thank you. Share on Twitter Share on Facebook This article is more than 4 months old Specifically, Brewer said teachers are angry that Republican leaders suspended a bipartisan working group that was examining potential changes to the pension system. They also are concerned about Senate bill 257. It adds gender neutral language to the state’s pension law, but Brewer said teachers worry that lawmakers will use it as a vehicle for a more expansive pensions overhaul.Kentucky’s pension systems are among the worst funded in the nation. State officials are at least $39bn short of the money required to pay benefits over the next 30 years.Lawmakers have made changes to the system that oversees state and local government employees. But they have been unable to make changes to the Kentucky Teachers Retirement System. Teachers are well organized and oppose anything they see as a threat to their profession.Thousands marched at the state capitol last year to protest against a Republican-backed bill that would have put all new teacher hires into a hybrid pension system. Lawmakers passed the bill anyway, but the state supreme court struck it down in December on a procedural issue.Thursday’s protests are likely to stir up the Democratic primary for governor, where several candidates are courting the support of teachers upset with the Republican-dominated state legislature and the Republican governor, Matt Bevin.Democratic candidate Adam Edelen called House bill 525 “political retribution against Kentucky’s teachers for having the guts to stand up to Matt Bevin and Frankfort politicians”. Another candidate, the attorney general, Andy Beshear, was on his way to the scene. Sign up for the US morning briefing Topics At least six school districts were forced to close as a wave of teachers called in sick to protest ‘half-truths’ and ‘shadiness’ Share on WhatsApp Last modified on Thu 28 Feb 2019 12.53 EST Kentucky Share on Twitter Share via Email US education This article is more than 4 months old Kentucky Share via Email At least six Kentucky school districts were forced to close on Thursday, including the two largest systems in the state, as a wave of teachers called in sick to protest what one leader called “half-truths” and “shadiness” at the state legislature.The “sick out” was organized by the group KY 120 United, a reference to the state’s 120 counties, which encouraged teachers in a Facebook post on Wednesday night to call in sick.Within hours, the state’s two largest school districts in Jefferson and Fayette counties announced they would close because they did not have enough teachers to cover classes. Districts in Marion, Carter, Letcher and Boyd counties were also closed on Thursday. In Fayette county, officials said at least 40% of teachers called in sick.The protest comes on the same day lawmakers in the Republican-controlled state legislature were scheduled to discuss House bill 525, a proposal that would change how people are nominated to the Kentucky Teachers Retirement System’s board of trustees. But Nema Brewer, one of the group’s leaders, said the protest is not targeting one specific bill.“This is [about] a series of kicks in the teeth. And the dog is now ready to fight back,” she said. Teachers from across Kentucky gather inside the state Capitol to rally for increased funding and to protest changes to their state funded pension system on 13 April 2018.Photograph: Bryan Woolston/AP Thu 28 Feb 2019 12.00 EST Reuse this contentlast_img read more

Assistive Technology and Autism Training November 19

first_img125 people registered to attend the Assistive Technology and Autism training on November 19.  It was a free training held at Easter Seals Crossroads for anyone interested in learning more about assistive technology devices for people of all ages with autism.Our guest speakers were:Scott Fogo, Director of Autism ServicesAndrew Goff, Assistive Technology SpecialistWendy Gahimer, Augmentative Communication SpecialistOverall the training was well received!  Many people came to learn about assistive technology because they are special education teachers, parents of children with autism, vocational rehabilitation counselors, or other professionals who work directly with people with autism.There were three assistive technology vendors present: Tobii ATI, Spec-Tech and Sensory Critters.  Each attendee was able to visit the vendors and receive informational brochures about INDATA’s serivces, our Assistive Technology and Transition from High School pamphlet, and resources from Prentke Romich Company‘s new AAC and Autism Center.  Also available were pamphlets about the Autism Family Resource Center, the Respite Program at Easter Seals Crossroads and an overview of autism services.INDATA provides four assistive technology trainings each year, and one is in conjunction with the annual assistive technology conference.  If you’d like more information on upcoming trainings, visit the Event Calendar or contact Sara Croft at scroft@eastersealscrossroads.org.Share this…TwitterFacebookPinterestLinkedInEmailPrint RelatedAutism and Assistive Technology TrainingOctober 9, 2009In “Conferences and Events”2013 INDATA Full Day Trainings Announced!January 9, 2013In “INDATA News”Assistive Technology and Cerebral Palsy Training, February 19March 2, 2010In “Conferences and Events”last_img read more

ATU306 – Unus Tactus – John Alliano Owner wwwunustactuscom

first_imgShare this…TwitterFacebookPinterestLinkedInEmailPrint RelatedATU200 – Wade Wingler is interviewed by Danny Wayne in This Special Celebration of 200 Episodes of Assistive Technology UpdateMarch 27, 2015In “Assistive Technology Update”ATU192 – Assistive Technology Lending Libraries (John Effinger), Tattos that Measure Blood Glucose, Microsoft’s HaloLens for People with Disabilities, Super Bowl Accessibility, Smartphones Helping or Hurting Deaf CultureJanuary 30, 2015In “Assistive Technology Update”ATU256 – ATLAS Vantagepoint, Podcasts Now on Google Play, Google AT Grants Awarded, $400k ALS AT Challenge PrizesApril 22, 2016In “Assistive Technology Update” Podcast: Play in new window | DownloadUnus Tactus – John Alliano, Owner | www.unustactus.com*Applause* YouTube’s caption upgrade shows how machine learning is helping the disabled http://buff.ly/2nKMm4uBrain-controlled muscle stimuation makes paralysed man move limbs again http://buff.ly/2o4klHNWebinars – Assistive Technology Industry Association http://buff.ly/2p1GlRaApp: Sounding Board | www.BridgingApps.org——-transcript follows ——JOHN ALLIANO:  This is John Alliano. I’m the owner of Unus Tactus. And this is your Assistance Technology Update.WADE WINGLER:  Hi, this is Wade Wingler with the INDATA Project at Easter Seals crossroads in Indiana with your Assistive Technology Update, a weekly dose of information that keeps you up-to-date on the latest developments in the field of technology designed to assist people with disabilities and special needs.Welcome to episode number 306 of Assistive Technology Update. It’s scheduled to be released on April 7, 2017.Today I have a fascinating conversation with John Alliano who is the owner and one of the cofounders of Unus Tactus, one of the apps it has recently won the FCC Chairman’s award. Also we have some information about something that you to be doing to improve its captioning service, thinks about adding sound effects, things like applause. And we have some information from ATIA on some of their upcoming webinars. We also have an app from BridgingApps called sounding board.We hope will check out our website at www.eastersealstech.com, we’d love for you to shoot us a note on Twitter at INDATA Project, or did you know you can leave a voicemail for any of our shows at our listener line 317-721-7124.***For a while now the FCC rules in the US have indicated that television stations have to provide closed captions that convey speech, sound effects, and audience reactions, things like laughter and applause. YouTube however hasn’t been subjected to those rules but has recently added some algorithms to their machine learning technology that is supposed to address that. Recently YouTube added some algorithms that can indicate applause, laughter, music in captions. There is some hint that they might also include things like door knocks, sighs, and barks in the future. My experience has been that captioning on YouTube isn’t always 100 percent accurate but seems to be getting better over time, and I’m glad to know they are investing resources and improving it with the stuff. Hopefully the overall captioning experience will continue to get better over time and we will have a flawless captioning system on YouTube. I found this information from MIT technology review that talks about some interesting mainstream technologies and accessibility, so I would encourage you to check out that article and I’ll pop a link in our show notes.***I spent a lot of time looking at social media. Just in the last several days I’ve seen tons of headlines about this story. This particular one is from the International Business Times, and the headline reads, “Braincontrolled muscle stimulation makes paralyzed man move limbs again.” Apparently William Cocovar has been a person with a spinal cord injury for about 10 years and now is using some new technology that allowed him to move. One of the researchers at Case Western Reserve University who is a biomedical engineer says that they are working to circumvent the spinal cord injury. They are using neural prostheses that connects his brain to his muscle. They say that he has been able to eat and drink again, some pretty significant movement there. Apparently there are two small recording chips that have been implanted into his motor cortex and then 36 electrodes embedded into his right arm to be able to do that narrow muscle stimulation. They say the end result is that he is able to coordinate moving and grasping movements in his own paralyzed arm and hand. This is pretty remarkable stuff that seems to be picking up a lot of media. I’m going to pop a link in the show notes over to the international business times post, but know that there is lots of information out there about this on the web. Check our show notes.***Are you interested in learning more about assistive technology? ATIA, the Assistive Technology Industry Association, has some webinars coming up, some that will still be available at the airing of this show. April 13 is “The iPad and total access for the low vision student in the regular classroom”, or April 19 “Creating communication competence for people with severe impairments.” These things tend to run about 330 to 5 PM Eastern time on those days. I’m going to pop a link in the show notes over to the ATIA website so you can learn more about what’s available. Check our show notes.***Each week one of our partners tells us what’s happening in the ever-changing world of apps, so here’s an app worth mentioning.AMY BARRY:  This is Amy Barry with BridgingApps and this is an app worth mentioning. This week’s app is called Sounding Board. Sounding Board transforms or iPhone, iPod touch, or iPad into the latest and communication technology. In just minutes, you can create custom boards using AbleNet symbols or your own words. Perfect for children in special education, people on the autism spectrum, and adults with disabilities. Sounding board includes switch access and is compatible with the AbleNet Bluetooth switch.The app is preloaded for free with 14 boards with the options of nine pictures per board depending on the needs of the user. AbleNet also has an extensive library of pictures to change photos and there is the option of using your own photos for certain boards. The other really nice feature is the option to use your own voice when editing pictures such as your name and phone number. Editing the board is very user-friendly. Caretakers, teachers, or therapists just turn off the edit locked in the settings and go to the board they want to edit and either add or delete pictures. This is also where you can change the given voice and put in your own.There is an option for in app purchases from AbleNet to buy boards dealing with restaurants, movies, general conversation, getting dressed, and mealtime. They are each separated by female or male, making the expressions more appropriate for a variety of user.This app is one of our tools that our reviewer uses in her classroom as it helps with twice making, expressive language, receptive language, and even social skills.Two important boards are the emergency information and emergency help. This is a must-have tool because it can be adjusted to the needs of the child. Our nonverbal students will use this app when they need to get information out quickly, especially for requesting help.Sounding board now allows auditory scanning in addition to single and dual switch scanning. Auditory scanning works by playing a short, prompt message during single or dual scanning actions. When the user selects the appropriate cell, the full message place.Sounding board is available for free at the iTunes Store and is compatible with iOS devices. For more information about this app and others like it, visit BridgingApps.org.***WADE WINGLER:  I’m always on the lookout for new apps. I particularly fascinated when I can find one that I think is innovative and does something that I haven’t seen done before. One of the ways I find out about apps and things like that as I pay attention to the FCC Chairman’s awards that are all about accessibility and helping folks with disabilities in the area of communication. Not long ago I was looking over some of the recent Chairman’s award winners and I saw a thing called Unus Tactus. I thought, I think that means my Latin lessons were failing me. I think it means one touch. I became immediately fascinated.In the conversations leading up to today’s interview, I got to know a guy named John Alliano who is a peach of a guy and is going to tell us a little bit about the story of Unus Tactus and what it is and who it helps and those kinds of things. Before I go on any further, John, are you still on the line?JOHN ALLIANO:  I certainly am. How are you doing?WADE WINGLER:  I’m doing great. Thank you so much for being with us today. Also thank you for rescheduling. I’ve been a little under the weather lately and we had to bump it around, but we finally had it scheduled and I’m glad we have a chance to talk.JOHN ALLIANO:  Never a problem. I truly appreciate the opportunity to speak to you and your audience.WADE WINGLER:  Thank you. Before we get into talking about the app, tell us a little bit about you and your background and then lead us into where the idea for Unus Tactus came from.JOHN ALLIANO:  For my background, I’m a retired police captain with 27 years in a municipal police department here in New Jersey. Through the course of my career, I had occasion to run into situations where folks were lost, wondered, children were missing and stuff like that. Afterwards my younger daughter, who is a certified speech pathologist, while she was in between her undergrad and her masters program, was working in a private practice as a secretary and had the opportunity to speak to various caregivers in a speech pathology private practice. She saw that it was extremely important for the people, the clients coming through this practice to be able to communicate properly. She found that through speaking with these caregivers that there was an issue with having to use cell phones and also the tracking devices. She spoke to me about it and said, “Dad, I was looking around to the Internet and there is no telephone out there that you can easily use to make phone calls.” She said, “The caregivers are telling me that they are embarrassed or they feel self-conscious about having to carry around trackers or where a tracker around their neck and such like that. What do we do.” I said, “Let’s develop something and do an app.” She said that would be a great idea. We move toward on it.WADE WINGLER:  I’m fascinated for lots of reasons. The business guy in me is interested in how you go from an idea to something that actually shows up on somebody’s phone. I guess I want to get into that before we talk about what it does and some of the features. How do you go from this conversation of, “That would be a cool app,” to “Here’s my cool app.” What does it look like?JOHN ALLIANO:  What we did is she and I sat down and talked about it and said, “What do you feel is important?”  She said, “Dad, the thing is that folks sometimes have an issue with remembering or recognizing the written word. A photograph they can make that identification. I would like to do a photo dialer.” I said, “I’m sure that’s something that’s relatively easy to do.” I said to her, “What are some of the other issues that these kinds of folks have?”  She said, “They have a tendency to wander or they have issues with having to hit buttons and stuff like that.” We came up with a format where we limited the number of dialing features on a screen so that it would be relatively easy to dial out, and then we also looked into the wandering problem and also the situation where folks would maybe get confused and forget where they were in later stages. We look at all these various situations. I don’t know how to do coding and neither does my daughter. We look for a good programmer or developer and found one in Princeton New Jersey. They are a great group of guys and were very excited. We spoke with the owner and he was very excited about it, because most of the time they do games. But to do something that would be actually helpful for folks was a fantastic idea in his estimation. We worked hand in hand and developed it.WADE WINGLER:  That sounds like a great family project and a good business obligation as well. I sort of put together in my mind a few things about what the app is and what it does. Walk me through the features. Tell me specifically what it does and what the beaches are and how it works.JOHN ALLIANO:  First of all, you can put an unlimited amount of contacts, and it can be stored through the phone and on the app. The app has up to four contacts per screen and you can scroll through the screen to get additional contacts. Each screen has a 4 x 4 grid so that it’s easy to see the particular pictures and touch the pictures. Let’s say for instance you went and put a contact in and didn’t have a picture yet. It’ll give you a great shade square outline until you get a picture so you can still make that call if you had to before you get the picture in place. The picture and name displays on every contact.And there is also a geo-fence feature. That geo-fence is an imaginary outline or border that you can designate from one mile to 15 miles from a central point they designate. That helps with the wandering problem or if you get lost or the phone or you end up in the wrong place. What that would do is it sends an emergency email message. I’ll get further into that further into the feature list.It also has a help button on the bottom. That help button can be independently programmed for any phone number. You can do it for a caregiver, or if the person is more independent and able to do more things, you can program that for 911. The matter where you are in the US, you hit that button and you will be locating the nearest 911 answering point and be able to get police and emergency services help very quickly.It has an optional email alert for every call made. What that means is you can program this app so that every time a person makes a phone call, the caregiver will get an email. On the email there is a Google map link. You hit that link on the email, it will show you exactly where the phone call is being made from. For instance, my father tells me, “John, I’m going to be going to such and such repair shop.” I tell them, “Okay dad, when you get there, give me a call and let me know you are there okay.” And he does and I find out that he is at the grocery store. Then I can say, “You know what?  Hang out there for a second. I’ll be right there. I need to do something over there also.” Now I’m not embarrassing him. I’m not calling the police and further embarrassing him with emergency services and he may get confused like why they are there. I’m de-escalating the situation and handling it with the least amount of disruption. That’s a great feature.Insofar as putting information in, you can either put in new contacts which is a very simple, or you can import them from your existing contacts. By the way, this app is designed for the iPhone right now. And also has the ability for an emergency contact with links to the iCloud. A scenario would be, let’s say for instance the phone were to leave the geo-fence area. You get a different type of email that would say that describe in the email that the phone has left the geo-fence area. And then there are two links that are connected to the email. One is to Google maps which shows you historical point of where the phone left the geo-fence. And then there is another link for iCloud. You hit that and type in your password and you can track the phone lives. The great thing about that is, let’s say for instance mom or dad or whoever is driving down the road, and they get lost and go outside the geo-fence. You call them, you can’t get through. You can actually call the police department and tell them turn by turn where that vehicle is so they can safely pull them over and de-escalate the situation before mom and dad panics or whatever that they are totally lost. You can be proactive. The time from which the phone call is made or the phone goes outside the geo-fence is about 30 seconds. It’s really quick. It’s just as fast as the technology will allow.We’ve had this out here, I’m going to say, about four years at least. It’s been sold all over the world. As a matter of fact, your listeners in Australia are some of the ones that bought the first apps. We were blown away because this is our first endeavor into this type of technology. For us to sell our first apps in Australia was quite awesome. We’ve done some great contacts with the folks out there.WADE WINGLER:  Excellent. We do have a pretty good audience in Australia so I would love to hear from most folks listening down under to let us know if you’ve had experience with it.JOHN ALLIANO:  That would be great.WADE WINGLER:  I’m starting to get a picture here of all the different features and how it works. Just because of my experience and probably true with my listeners, I’m starting to get a picture of the “who”, who is this for and what kind of disabilities are we talking about. But flesh that out a little bit for me. Tell me about the folks who might benefit from having his app on the phone.JOHN ALLIANO:  Folks that have Alzheimer’s or dementia, especially the very early diagnosed folks, there’s going to be that gradual tendency to wander in possibly forget where they are at from time to time. Sometimes I forget where I met. You get disoriented for a moment. That could be up to 20 months prior to the onset of serious situations where you are no longer fully independent and you need some additional physical assistance. If you have the onset of word recognition problems because it’s easier to recognize photos. If you have difficulty with multistep directions because of some memory loss. These are the types of situation and symptoms that the app will help with because you’re going to have photographs and words together which will help you recognize who you want to talk to. You have that geo-fence which helps with the wondering. You have the help button which is very visible, where you know if you hit that button you’ll be getting help.Just as importantly, it allows you the independence. I personally can’t imagine what it’s like to understand and realize that you are losing your independence. This is one of the reasons why this app was developed, is to extend that period of time where you have independence.This is a growing problem. Over the next 30 or so years, we are going to have an increased population that’s going to fall into these categories. You have people with cognitive deficits, folks on the autism spectrum, young folks that have similar types of symptoms. They can be helped. Children with dyslexia or people with dyslexia, it’s going to help them because you have a photo which is more easily recognized. I personally know people that have essential tremors. My wife has these essential tremors. I couldn’t understand just how difficult it is to make a phone call. By having these buttons, it’s just much easier to make that phone call. Even if your hand is shaking or not, it makes it that much easier. I use it every day. Thank God I don’t have any disabilities or restrictions. I like it because my hands are huge and I keep dialing the wrong number when I hit that button.WADE WINGLER:  I have big thumbs.JOHN ALLIANO:  It’s great for just regular use. The nice thing about the app is the tracking capability where if a parent has a new driver, they can set those parameters up. If a student is away at school, they can set it up. So long as both parties know what’s going on, it can be an assistive type of device where it gives everybody piece of mine. That’s important also for the caregivers, peace of mind knowing that your loved one is able to get help if they need it. That’s a huge plus. Visual defects, it’s obviously something that will help. If you have a vision problem, with the larger pictures and only four contacts per page cut it makes it easier to use. Blurred vision, it definitely helps with that. Seniors who are not really adept with technology — and I’m one of them and I’m okay. I am no techno-nerd. I can do this. To set this app up — and this is no exaggeration — takes less than five minutes. The geo-fence feature and the email feature have an on-off switch, because if it’s not necessary you don’t need to use it. But if you want to use it, it’s just a matter of going into the settings on the app. That was one of the important thoughts when we were thinking this out. We wanted to make it simple to utilize and to modify. We kept the features to a minimum to the most important features. That was key.I think we covered a lot of the various areas unless you have a specific question in so far as additional usage. Even young children, it’s great for working mothers and parents that want to say, “Johnny, you are going to go to such a place. Call me when you get at your friends house for homework or whatever.” They can make that phone call, get the email, and check on Google maps to make sure they are where they are supposed to be.WADE WINGLER:  It’s on like all the critical information is right there. We are running out of time for the interview, but before we finish up, tell folks about what it costs to get the app and where they get it. If they want to take action on this, how do they do that?JOHN ALLIANO:  You go to the App Store and type in Unus Tactus. The cost is $9.99 and is a one-time fee. There are no recurring charges. We just updated it for the newest operating system for Apple. You get those updates as you would for any application. There is no additional cost for that. It basically works all over the world.WADE WINGLER:  We talked about your daughter, the speech pathologist who helped with this, but we never said her name. What’s her name?JOHN ALLIANO:  Her name now is Ashley McCarthy. She got married a year ago in June 2016.WADE WINGLER:  Excellent.JOHN ALLIANO:  Ashley McCarthy is her new name and she’s a certified speech pathologist.WADE WINGLER:  Ashley, I hope you are listening. Your dad did a great job. John Alliano is the owner of Unus Tactus and has been our guest today explaining this cool app and all the things that it does. Thanks so much for being with us today.JOHN ALLIANO:  You’re welcome. Did you want the email and twitter?WADE WINGLER:  Let’s go ahead and have those!JOHN ALLIANO:  nusTactus@yahoo.com is the email. The twitter account is @UnusTactus.WADE WINGLER:  Thank you so much.JOHN ALLIANO:  Thank you.WADE WINGLER:  Do you have a question about assistive technology? Do you have a suggestion for someone we should interview on Assistive Technology Update? Call our listener line at 317-721-7124, shoot us a note on Twitter @INDATAProject, or check us out on Facebook. Looking for a transcript or show notes from today’s show? Head on over to www.EasterSealstech.com. Assistive Technology Update is a proud member of the Accessibility Channel. Find more shows like this plus much more over at AccessibilityChannel.com. That was your Assistance Technology Update. I’m Wade Wingler with the INDATA Project at Easter Seals Crossroads in Indiana.***Transcript provided by TJ Cortopassi.  For transcription requests and inquiries, contact tjcortopassi@gmail.com***last_img read more

Nvidia vs AMD Do GPUs Degrade in Performance Over Time

first_img Bran’s Deleted ‘GoT’ Scene Would Have Changed Everything By Joel Hruska on July 15, 2019 at 7:39 am The Radeon Vega 64 is the big winner here, climbing 1.14x, while Turing and Pascal both pick up about 1.11x. The GeForce GTX 1080 still gains 1.07x. The Radeon VII is the only GPU to show flat minimum performance.<><>123456 Tagged In amdgamingnvidiagpusPascalVegaturingGTX 1080RTX 2080Vega 64GTX 1080 Tirise of the tomb raiderCore i7-8086KRX 5700assassin’s creed originsshadow of the tomb raiderRX 5700 XTWindows 10 1903hitmanGigabyte Aorus RTX 2080 Xtremefar cry 5Deus Ex: Mankind DividedWarhammer IIWindows 10 1803Ashes of the Singularity EscalationMetro Last Light ReduxMiddle Earth: Shadow of War Post a Comment 98 Comments Don’t Watch These Movies With Your Parents One of the areas where reviews sometimes fall short is in providing perspective on how component performance evolves over time. While the yearly new hardware launch cadence provides periodic opportunities to revisit how older components stack up to newer ones, the focus in any given launch is going to be on the CPU or GPU that’s being launched, not previous cards. Title-specific coverage, meanwhile, is typically measured and written in the days immediately following a launch, often as part of a game or product review. Stack these two trends together, and it can be more difficult than it ought to be for gamers to figure out how performance has changed over time and to tease out which GPUs are holding their value better than others.With this in mind, we’ve completed an analysis of the data sets we gathered in our recent review of AMD’s Radeon RX 5700 and Radeon RX 5700 XT, as well as the Gigabyte Aorus RTX 2080 Xtreme 8GB. We updated all of our GPU data sets in late June / early July of 2019, which makes this a good time to look back at how Pascal, Turing, and GCN performance have each evolved over the past nine months.We’re specifically watching for two trends. First, there have been a number of general concerns from gamers about the ongoing impact of Meltdown and Spectre patches on game performance. Second, there is a perception in certain circles that Nvidia GPUs lose performance more quickly than their AMD counterparts, either because of intrinsic characteristics of Nvidia GPU design or because the company deliberately handicaps older cards to make newer GPUs look better by comparison.If I’m being honest, I’ve never believed the more sinister version of this argument. Nvidia and AMD pursued somewhat different optimization strategies in the pre-DX12 era, and Nvidia can reasonably be assumed to focus its optimization efforts on newer GPUs rather than older cards. This is not unique to Nvidia, however. Now that AMD has RDNA in-market, it may have to make decisions about how to prioritize its time when working to optimize for its various architectures as well. There is a difference between saying that Nvidia may focus more on optimizing for newer cards and saying that Nvidia deliberately handicaps older GPUs. In any event, the goal here is to measure how performance has evolved over time in the same suite of titles. We’ll see where the results take us.Test SetupAll of our tests were run on an Asus Prime Z370-A motherboard, with 32GB of DDR4-3200 using an Intel Core i7-8086K. The September 2018 Nvidia GPUs were tested using the 411.63 Turing launch driver, while the June retest used the 430.86 Nvidia driver. The AMD Radeon 64 and Radeon VII used the Adrenaline 19.5.2 driver. A Samsung 1TB 970 EVO was used for storage. The September 2018 tests were run using Windows 10 1803, while the June 2019 tests were run under Windows 10 1903. All Meltdown, Spectre, and related patches were left in default states.While the dates for the comparison are set to September 2018 and June 2019, this is obviously a bit of a fudge in the Radeon VII’s case (the Radeon VII didn’t launch until February). The Radeon VII’s June performance is being compared with its launch performance in this case.Two games showed performance declines across both Radeon and GeForce hardware: Ashes of the Singularity: Escalation and Warhammer II. Both games showed declines in all APIs, though AotS: Escalation lost more performance. This is theorized to be the result of Spectre et al protections. No other games showed performance declines, and the declines in these specific titles were not large enough to change the overall trend across our suite of games.We measured performance in Ashes of the Singularity: Escalation, Deus Ex: Mankind Divided, Hitman, Metro Last Light Redux, Middle Earth: Shadow of War, Rise of the Tomb Raider, Warhammer II, Shadow of the Tomb Raider, Assassin’s Creed: Origins, and Far Cry 5. The performance figures given for each GPU in each resolution reflect the geometric mean of our results. We used a geometric mean instead of an arithmetic mean to calculate averages to account for the fact that minimum frame rates can vary widely by game — Hitman, for example, regularly returns minimum frame rates for all GPUs between 4 – 12fps.Our September coverage used a standard GeForce RTX 2080, while the June 2019 data is based on the Gigabyte Aorus RTX 2080 Xtreme, which has slightly higher clock rates. This may have had a slight impact on performance (1-2 percent), but the difference is not large enough to be an issue for our purposes.ResultsThe slideshow below contains our results, graphed by resolution and by minimum versus average frame rates. You Might Also LikePowered By ZergNet Marvel Cinematic Universe Phase 5 Is Already Planned Conclusion:The minimum frame rate improvements at 1080p and 1440p are quite solid for Vega 64, the RTX 2080, and the GTX 1080 Ti. The average level of improvement is smaller across the board for all cards, but this isn’t necessarily surprising. Vega 64 and Radeon VII are both based on GCN, and GCN has been AMD’s lead architecture for a number of years — long enough to be well-optimized, at this point. Nvidia’s RTX 2080 picks up the most consistent gains across all resolutions, probably thanks to the minor clock tap or the fact that Turing is the newest architecture and had the most performance still on the table. Even the GTX 1080 picks up a few frames at 1080p.There is no evidence that Nvidia has taken any action to harm Pascal performance or to make its older, 2016 GPUs look less attractive than more recent cards. There are no Pascal performance regressions or frame rate issues in any game that do not affect every GPU (that’s why we think the performance decline in Ashes and Warhammer II is CPU-related, not GPU-related). Vega 64’s performance may have come up the most, but Vega 64 has been on the market for less time than the Pascal family. Despite these improvements, it also only matches or slightly underperforms the GTX 1080 at every resolution and in both minimum and average frame rates. The relative performance of these AMD and NV cards relative to each other has shifted only modestly.The implication of these findings is only to the good, no matter which GPU you own. AMD or Nvidia, Vega and Pascal cards both show continued expected performance, while Turing’s slightly larger improvements are what we’d expect for a relatively newer architecture. Obviously the particulars of how a GPU architecture ages are going to be specific to each architecture, and since this review didn’t focus on older cards from the Maxwell or Kepler era, we can’t speak to the situations with those GPUs. But as far as Pascal is concerned, Nvidia’s last-generation architecture appears to be aging quite well, while AMD has also improved on Vega’s performance.Now Read:AMD’s RX 5700 and RX 5700 XT Reviewed: RDNA Puts Radeon Back in the GameNvidia RTX 2060, 2070 Super: Better GPU Performance, Same PriceMeet RDNA: AMD’s Long-Awaited New GPU Architecture 98 Comments Facebook Twitter Linkedin Pinterest Google Plus Reddit Hacker News Flipboard Email Copy 1shares This site may earn affiliate commissions from the links on this page. Terms of use. Maisie Williams’ Transformation is Turning Heads Lilly From ‘Princess Diaries’ Is 36 Now and Gorgeous Marvel Fans Are Hoping Valkyrie Finds Her Queen Things Only Adults Notice in ‘ThunderCats’ Nvidia vs. AMD: Do GPUs Degrade in Performance Over Time? Willow Smith’s Transformation is Turning Headslast_img read more

Scientists Simulate Marsquakes With Data From InSight Lander

first_img 0 Comments You Might Also LikePowered By ZergNet This Movie Will Have More Fight Scenes Than Any Other MCU Film If you’ve ever wondered how different marsquakes are from earthquakes, we now have the answer thanks to NASA and Swiss research university ETH Zurich. NASA delivered the first-ever seismometer to the red planet aboard the InSight lander last year, and it’s now deployed and sending back data. Researchers from ETH Zurich have used that data to reconstruct a marsquake here on Earth, and it’s not what anyone expected. InSight touched down on Mars back in November of 2018. The robotic lander’s goal is to study the interior of Mars, which has not been possible with past missions like Curiosity. Among InSight’s instruments is the Seismic Experiment for Interior Structure (SEIS). NASA successfully deployed the sensor in December of 2018 (see above) after building a model of InSight’s landing zone on Earth to rehearse the delicate process. InSight placed the instrument about 5.3 feet (1.6 meters) away, close to the maximum reach of its robotic arm. If SEIS didn’t make good contact with the ground, its data wouldn’t be useful. Luckily, the sensor works perfectly and detected its first marsquake in April 2019. In a new video, ETH Zurich scientists demonstrate the differences between seismic events on Earth, the moon, and Mars. They have a quake simulator that can reproduce a seismic event from data as you’d get from an instrument like SEIS or the seismometer deployed on the moon by Apollo 11 50 years ago. Seismic waves are affected by the composition of a planet or moon, so you’d expect different worlds to have different quakes. Indeed, quakes on Earth feature strong initial P-wave arrival followed by a stronger S-wave. To experience moonquakes, the team had to scale the signal up by one million times. Seismic waves on the moon have much smaller attenuation with strong scattering. So a moonquake could last an hour or more. For the marsquake simulation, the team had to increase the InSight signal by 10 million times. The lander has so far detected two “families” of marsquakes. There’s a high-frequency version and the low-frequency version demonstrated in the video. The lower-frequency quakes may originate farther from InSight. The waves aren’t the same as on Earth, but they’re not like the moon either. ETH Zurich researchers believe Mars will end up having seismic properties in between the two. In time, the seismic data from Mars could help us better understand its composition without cracking it open to take a peek.Now read:InSight Lander Records the Sound of Martian WindInSight Takes Its First Mars SelfieInSight Beams Back Daily Weather Reports From Mars Facebook Twitter Linkedin Pinterest Google Plus Reddit Hacker News Flipboard Email Copy 0shares This site may earn affiliate commissions from the links on this page. Terms of use. New TV Shows That Will Absolutely Get Canceled in 2019 Tagged In sciencespacenasamarseth zurichseismologyseis Post a Comment 0 Comments By Ryan Whitwam on July 23, 2019 at 1:54 pm ‘Endgame’ Directors Apologize To Tom Holland We Finally Know Why This Kid Returned in ‘Endgame’ Deleted ‘Endgame’ Scene Showed Us Tony Stark In The Afterlife Scientists Simulate Marsquakes With Data From InSight Lander Don’t Watch These Movies With Your Parents Why Oliver Won’t Be Miserable In ‘Arrow’ Season 8 Winona Ryder Is Not Letting Anyone Forget She’s Married to Keanulast_img read more

Hubble Spots a Black Hole That Shouldnt Exist

first_img Facebook Twitter Linkedin Pinterest Google Plus Reddit Hacker News Flipboard Email Copy 0shares This site may earn affiliate commissions from the links on this page. Terms of use. You Might Also LikePowered By ZergNet Why Billy Hargrove from ‘Stranger Things’ Looks So Familiar ‘GTA’ Myths That Turned Out to be True Hubble Spots a Black Hole That Shouldn’t Exist Willow Smith’s Transformation is Turning Heads By Ryan Whitwam on July 13, 2019 at 8:20 am Anna Kendrick’s Transformation is Turning Heads Conventional wisdom holds that most large galaxies harbor supermassive black holes at their centers. Scientists also believe that only so-called “active” galaxies should have visible accretion disc of matter, but the Hubble Space Telescope has found one around a black hole with unusually low luminosity. This galaxy might bend the rules a bit, but it offers an opportunity to study how the theory of relativity applies in the real world. NGC 3147 is a large spiral galaxy just a bit smaller than our own Milky Way. It sits about 120 million light-years away — you’ve probably seen pictures of it because it’s quite stunning. Active galaxies like quasars are easy to spot. The matter falling into them produces emissions across the electromagnetic spectrum, and the accretion discs are quite visible. Everyone thought NGC 3147 was far too dim to have a disc of its own, but a new analysis from an international team suggests otherwise. Hubble collected data from the central black hole in NGC 3147, which has a mass about 250 million times greater than our sun. The object turns out to have a thin disc of material similar to what you’d find around an active galactic nucleus. Observations from Hubble show the disc spins at about 10 percent the speed of light. Researcher Stefano Bianchi from the Università degli Studi Roma Tre in Italy says this discovery indicates the current models for low luminosity galaxies have “clearly failed.”This discovery could have far-reaching effects on the way we model galaxies and black holes, but it also could lead to a better understanding of the physics underlying supermassive objects. The fast-moving disc combined with the low apparent brightness of NGC 3147 could make it possible to test both general and special relativity. General relativity deals with the mechanisms of gravity in the universe, and special relativity describes the relationship between space and time. Since NGC 3147 is dimmer than a typical active galaxy, we can make out the accretion disc around it much better. The disc is inside the black hole’s powerful gravity field where scientists can study the way it affects light. Based on findings in NGC 3147, astronomers may now go on the hunt for other weakly active galaxies. Some may even be closer to Earth where we can make even more precise measurements to test Einstein’s theories.Now read:Cold Quasars Could Change Our Understandings of Galactic DeathAstronomers Capture Historic First Photo of Black HoleResearchers Use Supersonic Fluid to Test Hawking’s Black Hole Theories ‘Jay and Silent Bob Reboot’ Trailer Packed with Hollywood Greats 13 Comments Things Only Adults Notice in ‘ThunderCats’ Sun Baby From ‘Teletubbies’ Is 22 Now & Unrecognizably Gorgeous The Tragedy of Marie Osmond Just Keeps Getting Sadder and Sadder Tagged In sciencespaceastronomyphysicsblack holeshubblegalaxiessupermassive black holes Post a Comment 13 Commentslast_img read more

How to Create Your Own StateoftheArt Text Generation System

first_imgHow to Create Your Own State-of-the-Art Text Generation System New TV Shows That Will Absolutely Get Canceled in 2019 The Transformation of Daisy Ridley Is Turning Heads Movies That Are Hard To Believe Actually Exist Winona Ryder Is Not Letting Anyone Forget She’s Married to Keanu You Might Also LikePowered By ZergNet By David Cardinal on June 26, 2019 at 12:01 pm Don’t Watch These Movies With Your Parents Little Lucy From ‘Narnia’ is Head-Turningly Gorgeous Now at 23 4 Comments 10 Scientifically Proven Side Effects of Growing a Beard Facebook Twitter Linkedin Pinterest Google Plus Reddit Hacker News Flipboard Email Copy 1shares This site may earn affiliate commissions from the links on this page. Terms of use. Hardly a day goes by when there isn’t a story about fake news. It reminds me of a quote from the favorite radio newsman from my youth, “If you don’t like the news, go out and make some of your own.” OpenAI’s breakthrough language model, the 1.5 billion parameter version of GPT-2, got close enough that the group decided it was too dangerous to release publicly, at least for now. However, OpenAI has now released two smaller versions of the model, along with tools for fine-tuning them on your own text. So, without too much effort, and using dramatically less GPU time than it would take to train from scratch, you can create a tuned version of GPT-2 that will be able to generate text in the style you give it, or even start to answer questions similar to ones you train it with.What Makes GPT-2 SpecialGPT-2 (Generative Pre-Trained Transformer version 2) is based on a version of the very powerful Transformer Attention-based Neural Network. What got the researchers at OpenAI so excited about it was finding that it could address a number of language tasks without being directly trained on them. Once pre-trained with its massive corpus of Reddit data and given the proper prompts, it did a passable job of answering questions and translating languages. It certainly isn’t anything like Watson as far as semantic knowledge, but this type of unsupervised learning is particularly exciting because it removes much of the time and expense needed to label data for supervised learning.Overview of Working With GPT-2For such a powerful tool, the process of working with GPT-2 is thankfully fairly simple, as long as you are at least a little familiar with Tensorflow. Most of the tutorials I’ve found also rely on Python, so having at least a basic knowledge of programming in Python or a similar language is very helpful. Currently, OpenAI has released two pre-trained versions of GPT-2. One (117M) has 117 million parameters, while the other (345M) has 345 million. As you might expect the larger version requires more GPU memory and takes longer to train. You can train either on your CPU, but it is going to be really slow.The first step is downloading one or both of the models. Fortunately, most of the tutorials, including the ones we’ll walk you through below, have Python code to do that for you. Once downloaded, you can run the pre-trained model either to generate text automatically or in response to a prompt you provide. But there is also code that lets you build on the pre-trained model by fine-tuning it on a data source of your choice. Once you’ve tuned your model to your satisfaction, then it’s simply a matter of running it and providing suitable prompts.Working with GPT-2 On Your Local MachineThere are a number of tutorials on this, but my favorite is by Max Woolf. In fact, until the OpenAI release, I was working with his text-generating RNN, which he borrowed from for his GPT-2 work. He’s provided a full package on GitHub for downloading, tuning, and running a GPT-2 based model. You can even snag it directly as a package from PyPl. The readme walks you through the entire process, with some suggestions on how to tweak various parameters. If you happen to have a massive GPU handy, this is a great approach, but since the 345M model needs most of a 16GB GPU for training or tuning, you may need to turn to a cloud GPU.Working with GPT-2 for Free Using Google’s ColabFortunately, there is a way to use a powerful GPU in the cloud for free — Google’s Colab. It isn’t as flexible as an actual Google Compute Engine account, and you have to reload everything each session, but did I mention it’s free? In my testing, I got either a Tesla T4 or a K80 GPU when I initialized a notebook, either one of which is fast enough to train these models at a reasonable clip. The best part is that Woolf has already authored a Colab notebook that echoes the local Python code version of gpt2-simple. Much like the desktop version, you can simply follow along, or tweak parameters to experiment. There is some added complexity in getting the data in and out of Colab, but the notebook will walk you through that as well.Getting Data for Your ProjectNow that powerful language models have been released onto the web, and tutorials abound on how to use them, the hardest part of your project might be creating the dataset you want to use for tuning. If you want to replicate the experiments of others by having it generate Shakespeare or write Star Trek dialog, you can simply snag one that is online. In my case, I wanted to see how the models would do when asked to generate articles like those found on ExtremeTech. I had access to a back catalog of over 12,000 articles from the last 10 years. So I was able to put them together into a text file, and use it as the basis for fine-tuning.If you have other ambitions that include mimicking a website, scraping is certainly an alternative. There are some sophisticated services like ParseHub, but they are limited unless you pay for a commercial plan. I have found the Chrome Extension Webscraper.io to be flexible enough for many applications, and it’s fast and free. One big cautionary note is to pay attention to Terms of Service for whatever website you’re thinking of, as well as any copyright issues. From looking at the output of various language models, they certainly aren’t taught to not plagiarize.So, Can It Do Tech Journalism?Once I had my corpus of 12,000 ExtremeTech articles, I started by trying to train the simplified GPT-2 on my desktop’s Nvidia 1080 GPU. Unfortunately, the GPU’s 8GB of RAM wasn’t enough. So I switched to training the 117M model on my 4-core i7. It wasn’t insanely terrible, but it would have taken over a week to make a real dent even with the smaller of the two models. So I switched to Colab and the 345M model. The training was much, much, faster, but needing to deal with session resets and the unpredictability of which GPU I’d get for each session was annoying.Upgrading to Google’s Compute EngineAfter that, I bit the bullet, signed up for a Google Compute Engine account, and decided to take advantage of the $300 credit Google gives new customers. If you’re not familiar with setting up a VM in the cloud it can be a bit daunting, but there are lots of online guides. It’s simplest if you start with one of the pre-configured VMs that already has Tensorflow installed. I picked a Linux version with 4 vCPUs. Even though my desktop system is Windows, the same Python code ran perfectly on both. You then need to add a GPU, which in my case took a request to Google support for permission. I assume that is because GPU-equipped machines are more expensive and less flexible than CPU-only machines, so they have some type of vetting process. It only took a couple of hours, and I was able to launch a VM with a Tesla T4. When I first logged in (using the built-in SSH) it reminded me that I needed to install Nvidia drivers for the T4, and gave me the command I needed.Next, you need is to set up a file transfer client like WinSCP, and get started working with your model. Once you upload your code and data, create a Python virtual environment (optional), and load up the needed packages, you can proceed the same way you did on your desktop. I trained my model in increments of 15,000 steps and downloaded the model checkpoints each time, so I’d have them for reference. That can be particularly important if you have a small training dataset, as too much training can cause your model to over-fit and actually get worse. So having checkpoints you can return to is valuable.Speaking of checkpoints, like the models, they’re large. So you’ll probably want to add a disk to your VM. By having the disk separate, you can always use it for other projects. The process for automatically mounting it is a bit annoying (it seems like it could be a checkbox, but it’s not). Fortunately, you only have to do it once. After I had my VM up and running with the needed code, model, and training data, I let it loose. The T4 was able to run about one step every 1.5 seconds. The VM I’d configured cost about $25/day (remember that VMs don’t turn themselves off; you need to shut them down if you don’t want to be billed, and persistent disk keeps getting billed even then).To save some money, I transferred the model checkpoints (as a .zip file) back to my desktop. I could then shut down the VM (saving a buck or two an hour), and interact with the model locally. You get the same output either way because the model and checkpoint are identical. The traditional way to evaluate the success of your training is to hold out a portion of your training data as a validation set. If the loss continues to decrease but accuracy (which you get by computing the loss when you run your model on the data you’ve held out for validation) decreases, it is likely you’ve started to over-fit your data and your model is simply “memorizing” your input and feeding it back to you. That reduces its ability to deal with new information.Here’s the Beef: Some Sample Outputs After Days of TrainingAfter experimenting on various types of prompts, I settled on feeding the model (which I’ve nicknamed The Oracle) the first sentences of actual ExtremeTech articles and seeing what it came up with. After 48 hours (106,000 steps in this case) of training on a T4, here is an example:The output of our model after two days of training on a T4 when fed the first sentence of Ryan Whitwam’s Titan article. Obviously, it’s not going to fool anyone, but the model is starting to do a decent job of linking similar concepts together at this point.The more information the model has about a topic, the more it starts to generate plausible text. We write about Windows Update a lot, so I figured I’d let the model give it a try:The model’s response to a prompt about Windows Update after a couple of days of training.With something as subjective as text generation, it is hard to know how far to go with training a model. That’s particularly true because each time a prompt is submitted, you’ll get a different response. If you want to get some plausible or amusing answers, your best bet is to generate several samples for each prompt and look through them yourself. In the case of the Windows Update prompt, we fed the model the same prompt after another few hours of training, and it looked like the extra work might have been helpful:After another few hours of training, here is the best of the samples when given the same prompt about Microsoft Windows.Here’s Why Unsupervised Models are So CoolI was impressed, but not blown away, by the raw predictive performance of GPT-2 (at least the public version) compared with simpler solutions like textgenrnn. What I didn’t catch on to until later was the versatility. GPT-2 is general purpose enough that it can address a wide variety of use cases. For example, if you give it pairs of French and English sentences as a prompt, followed by only a French sentence, it does a plausible job of generating translations. Or if you give it question-and-answer pairs, followed by a question, it does a decent job of coming up with a plausible answer. If you generate some interesting text or articles, please consider sharing, as this is definitely a learning experience for all of us.Now Read:Google Fed a Language Algorithm Math Equations. It Learned How to Solve New OnesIBM’s resistive computing could massively accelerate AI — and get us closer to Asimov’s Positronic BrainNvidia’s vision for deep learning AI: Is there anything a computer can’t do? This is Where ‘Thor 4’ Will Take Place in the Marvel Timeline Tagged In googleartificial intelligenceextremetech explainsdevelopersHow-Toprogrammingtensorflowpythonopenaifake newsgoogle compute engineGoogle ColabGPT-2text generation Post a Comment 4 Commentslast_img read more

Brand Networks Unveils FirstEver Solutions to Support Scale Authentic Local Social Advertising

first_img Brand NetworksNewssocial advertising Previous ArticleDiscoverOrg Survey Indicates Competition to Grow Dramatically Among U.S. CompaniesNext ArticleOpenSlate Launches YouTube Brand Safety Solution In Six Additional Languages Brand Networks, the cross-channel managed media software and services provider, announced a suite of local social advertising solutions to help brick-and-mortar brands reach, engage, and sell to their local communities in an authentic, credible and extensible way. Brands of all sizes can now generate local awareness, emotional connections, and store footfall with these solutions, leveraging and enhancing relationships with patrons.“Brick-and-mortar businesses are under attack in today’s digital world. They are looking for new ways to leverage their community presence to build local relationships,” says Dave Fall, CEO of Brand Networks. “Store managers and employees live in the community. They should be the ones localizing content so it resonates. However, optimizing social ad campaigns to maximize performance is an unreasonable requirement for in-store staff. With Brand Networks’ automation, a single digital advertising professional can operate thousands of local digital campaigns all running simultaneously.”Brand Networks is already powering local advertising at 5,000 brick and mortar locations nationwide. The value of these solutions doesn’t start and end with large enterprises either. Boutique brands like luxury leather retailer Schott NYC have also felt their effects.Marketing Technology News: Mastercard Digital Wellness Program to Enhance Transparency, Security and Choice for Online Shopping“The success of our new flagship store in Chicago was largely predicated on whether we could connect with local Chicago shoppers. Social advertising was a great place to start,” said Jennifer Goldszer, Vice President of Marketing for Schott NYC. “Brand Networks’ local social capabilities allowed us to connect our store and employees with the surrounding community, helping us create real content from real people while making the best use of every media dollar. In light of the pilot’s success, and knowing that today’s brands must authentically connect with consumers to remain competitive, it’s a strategy we plan to replicate across store locations.”With Brand Networks, Schott NYC took a hyper-focused approach to audience targeting and brought their new store to life through daily updates on Facebook and Instagram that keyed on local moments, local events, relevant hashtags, and surrounding businesses. Compared to the previous month, the hyper-local strategy led to a 27% increase in average weekly traffic at the store — and, impressively, a 111% increase in average weekly sales.Marketing Technology News: PROS Launches Sales Agreement Management to Streamline Selling in Digital EraThe new Brand Networks’ suite offers five core capabilities for local social advertising efforts:Local Inventory Ads: Control SKU promotion on a regional, local, or national level while adding or removing items on the fly; automatically pause ads when inventory runs out to reduce customer friction and eliminate wasted impressions.Locally Sourced, Relevant Content: Empower local employees to promote in-store events; respond to weather or news-related specials; give local shoutouts to build community, or profile stellar staff. Regional or corporate brand managers still retain oversight and approvals while incentivizing local employees to live the brand out loud.Local Audience Activation: Set radius targeting parameters to reach in-market customers on a per-location basis. Extend campaign reach or focus by leveraging first-party audiences and lookalike modeling to find ideal prospects for local campaigns.Local Measurement: Achieve a holistic view of online performance across stores through unified reporting and unlock location-level insights to inform future campaigns. Measure the impact of digital campaigns on foot traffic and offline purchases.Local Support & Professional Services: To ensure adoption, Brand Networks trains local teams, customizes tools and technology, and provides strategic and technical media management, empowering advertisers from setup to execution, from reporting to insights derivation.Marketing Technology News: Agency Veteran Joao Machado Joins as Sabio’s SVP of Product Marketing Brand Networks Unveils First-Ever Solutions to Support & Scale Authentic Local Social Advertising Globe NewswireJune 13, 2019, 1:56 pmJune 13, 2019 last_img read more

Acorn International Establishes Acorn Digital Services

first_img AcornAcorn Digital ServicesAcorn Entertainmentmarketing and brandingMarketing TechnologyNewsSocial Media Management Previous ArticleHow AdsDax Is Leveraging Blockchain in a Bid to Defeat Ad FraudNext ArticleOracle Ushers in New Era of Analytics Social Media and Digital Services Agency under Acorn International to Target Broader Range of Brands Seeking to Maximize Their Business in ChinaAcorn International, Inc. (“Acorn” or the “Company”), a leading marketing and branding company in China, announced that it has consolidated under Acorn Digital Services the following business units: Acorn Entertainment (“AE”) and Acorn Streaming, which are Acorn’s social media management business and digital content business, respectively, as well as A-KOL (standing for “Acorn Key Opinion Leader), which facilitates the development of local China influencers with the objective to optimize their ability to convert Acorn’s content into product sales.Acorn International’s business model combines social media management, content creation and sales of its own as well as third-party brands through e-commerce. The establishment of Acorn Digital Services over Acorn Entertainment, Acorn Streaming, and A-KOL consolidates Acorn’s digital agency business under one unit.Marketing Technology News: Global Software Leaders Xero and Stripe Join Forces to Bring Seamless Payments to Millions of Small BusinessesMr. Jacob A. Fisch, Acorn’s President and CEO noted: “Acorn’s business is and has always been about incubating promising new brands in a low risk manner. Acorn Digital Services is structured to allow us to incubate these brands while getting paid as we go as a third-party service provider.”In regard to the overall business model, Mr. Fisch noted further: “Social commerce is at the intersection of social media, content creation and e-commerce, allowing brands of all sizes to increase conversion rates by leveraging their social media reach. Brands that are embracing this rapidly evolving trend in China are getting a head start on the competition and positioning themselves for success. We believe Acorn Digital Services’ expanded focus will result in greater visibility and higher conversion rates for Acorn and our clients in China.”Marketing Technology News: GoSpotCheck Gains Valuable Leadership Expertise with Five Additions to Management TeamAcorn International is a leading marketing and branding company in China, leveraging a twenty-year direct marketing history to monetize brand IP, content creation and distribution, and product sales, through digital media in China. Marketing Technology News: Whitebox Raises $5 Million in Series A Funding to Accelerate eCommerce “Factory Floor to Front Door” Tech Platform Acorn International Establishes Acorn Digital Services PRNewswireJune 26, 2019, 2:09 pmJune 26, 2019 last_img read more

Salsify Announces New GDSN Data Pool Designed for the Digital Shelf

first_img GDSNMarketing Technology NewsMaster Data ManagementNewsProduct experience managementSalsify Previous ArticleRackspace and Tech Mahindra Partner to Lead with New Go-To-Market StrategyNext ArticleQualtrics Ranked Top 10 Most Innovative Company in 2019 GRIT Report Brand Manufacturers Can Manage and Deliver Content for Retail and Ecommerce Channels from One PlatformSalsify, a product experience management (PXM) platform that helps brands win on the digital shelf, announces a new data pool for GS1’s Global Data Synchronization Network (GDSN). This expanded capability will allow IT, Master Data Management (MDM), ecommerce, trade, and marketing teams to manage and share their product data with retailers and distributors across the GDSN from a single platform.As the only PXM platform with both product information management (PIM) capabilities and a native data pool that covers GS1’s complete Global Data Dictionary (GDD), Salsify now provides customers with an end-to-end solution for managing and transmitting product content for both traditional wholesale – where GDSN has been vital – and ecommerce channels. The Salsify platform unites the work of all teams involved in bringing products to market, streamlining processes, confirming responsibilities and handoffs through integrated workflows, and providing a single source of truth for all product data. This unified solution results in:Marketing Technology News: UiPath Named a Leader in the 2019 Gartner Magic Quadrant for Robotic Process AutomationFaster item setup: The unification of both logistical and top-off marketing data in one platform significantly reduces the time it takes to bring a new product to market.Content refresh at the speed of commerce: Optimizing for SEO, publishing new packaging variations, adding more engaging media such as 360 spins and lifestyle images, and reacting to consumer feedback in ratings and reviews all require frequent updates to product detail pages and product content. Bringing ecommerce teams into the loop with Salsify’s unified solution ensures that rich content that changes frequently can go through the same GDSN that supply chain data does.Compliance with confidence: Salsify maintains the complex requirements for each region and target market, enabling customers to stay compliant with global retailer requirements with a single data pool. The platform’s readiness reports and workflows immediately surface violations and users can assign tasks for remediation.Marketing Technology News: JumpCrew Announces Inaugural Conference “JumpCon: The Digital Sales Transformation Summit”“Our customers have pushed us to unite their offline and online product content processes to serve the need for accurate, complete, and differentiated product experiences everywhere on the digital shelf,” said Jason Purcell, co-founder and CEO of Salsify. “Adding a GDSN data pool to our platform helps bring to life GS1’s original vision of efficient and automated exchange of continuously refreshed product data across the globe.”The Salsify for GDSN data pool is currently in the certification process with GS1. Certification is expected this quarter, with immediate availability thereafter.Marketing Technology News: Valassis’ Marketing Technology Platform Earns Tech Award Salsify Announces New GDSN Data Pool Designed for the Digital Shelf PRNewswireJuly 12, 2019, 2:44 pmJuly 12, 2019 last_img read more

Study from Project Management Institute Identifies Six AI Technologies Impacting Project Professionals

first_imgStudy from Project Management Institute Identifies Six AI Technologies Impacting Project Professionals Business WireJune 19, 2019, 2:40 pmJune 19, 2019 AIdeep learningmachine learningMarketing TechnologyMichael DePriscoNewsProject Management InstituteRPA Previous ArticleQumu Signs International Partnership Agreement with London Startup CaptionHubNext ArticleTechTarget Priority Engine Named Best Account Based Marketing and Sales & Marketing Intelligence Solution by SIIA Project Management Institute (PMI) released its 2019 Pulse of the Profession® In-Depth Report: AI Innovators: Cracking the Code on Project Performance. The report provides an in-depth look at how artificial intelligence (AI) is impacting organizations and the project management profession. Findings reveal that AI disruption is happening and at a large scale: 81 percent of respondents report their organization is being impacted by AI technologies; 37 percent say adopting AI technologies is a high priority for their organization; and project professionals say they expect the proportion of the projects they manage using AI will jump from 23 to 37 percent over the next three years.Marketing Technology News: Databricks Accelerates APJ Expansion Following $250 Million Funding RoundThese insights from a survey of 551 project management practitioners globally show that the presence of AI technologies will continue to grow, requiring shifts in how projects are managed and how organizations implement strategy. The report identifies six AI technologies that are impacting organizations globally. The three AI technologies currently impacting the greatest number of organizations are:Knowledge-based systems: Understands the context of the data being processed, helping support human learning and decision making.Machine learning: Analyzes data to build models by detecting patterns, yielding improved decision making with minimal human intervention.Decision management: Creates an intelligent process or set of processes based on rules and logic to automate decision making.Over the next three years, project professionals expect the impact of these AI technologies to grow:Expert systems: Emulates and mimics human intelligence, skills or behavior in a particular field, topic or skill.Deep learning: Builds, trains and tests neural networks that predict outcomes and/or classify unstructured data based on probabilities.Robotic process automation: Mimics and automates human tasks to support corporate processes.In a recent Pulse of the Profession report, PMI called for organizations to bolster their Project Management Technology Quotient (PMTQ), which is a way of evaluating an organization’s ability to manage and integrate technology based on the needs of the organization or the project at hand. As organizations increasingly adopt AI technologies, having a strong PMTQ will be essential.“What this report shows is that the time is now to start thinking about how organizations can leverage AI technologies to achieve project success,” said Michael DePrisco, vice president of global solutions at PMI. “Since project professionals are largely responsible for helping to implement change, they will play a critical role in influencing the adoption of these technologies in their organizations. Based on PMI research, we know that organizations with a high Project Management Technology Quotient (PMTQ) understand that capturing the value of AI will take more than simply rolling out the technology. It will require a new mindset, a learning culture and support from the top to establish an organizational infrastructure that leads to success.”Marketing Technology News: Blis Expands Into the Netherlands With First HireIn fact, research by Accenture finds that most visionary organizations apply five key principles to their AI investments. Called the MELDS principles, they call on organizations to:Encourage the right MindsetPromote ExperimentationSupport actively involved LeadershipIncorporate Data into strategyCultivate new SkillsOrganizations who embrace these five principles are called AI Innovators and are more likely than organizations that don’t embrace any of the principles (AI Laggards) to establish an organizational infrastructure that leads to project success. According to Pulse® data, there is a distinct payoff for organizations leading the way with AI. AI Innovators outperform AI Laggards on several key project metrics:Better on-time delivery: AI Innovators report they delivered 61 percent of their projects on time, versus 47 percent for AI Laggards.Superior benefits realization: AI Innovators report 69 percent of their projects realized 95 percent or more of their business benefits, compared to 53 percent of projects for AI Laggards.Higher ROI: AI Innovators report 64 percent of their projects met or exceeded their original ROI estimates, versus 52 percent of projects for AI Laggards.Successful implementation of AI technology requires organizations to shift their priorities. Not only is it important to look at PMTQ as a measure of preparedness, but also how well an organization adheres to the MELDS principles. This will be especially important as more and more AI technologies are integrated into operations in the coming years.Marketing Technology News: Shoppers Take Center Stage in the 2019 Retail Systems Research Report on eCommerce Website Performancelast_img read more

Very Clear BJP Govt is Going Says Priyanka Gandhi After Casting Vote

first_imgNew Delhi: Congress leader Priyanka Gandhi Sunday said it was “very clear” that the BJP will face defeat in Lok Sabha polls as people are “angry and distressed” and will convey their sentiments through voting.Interacting with reporters after exercising her franchise, the senior party leader said, it was “very clear that the BJP government is going”. “There is anger among people, and they are under distress. Modiji, instead of talking about real issues, has been talking about random things. And, now, they will express their anger against this government through voting,” the Congress general secretary said.She said this will be witnessed especially in Uttar Pradesh.Priyanka again hit out at Prime Minister Narendra Modi, and said, “He doesnt reply to any of the issues that our party raised, and just talks about random things.”Congress campaigns on real issues and the prime minister has “not even responded to the challenge thrown by our party president (Rahul Gandhi) to debate on real issues,” she said.He and his party had in 2014 promised to deposit Rs 15 lakh in every Indian’s bank account and create two crore jobs, but Prime Minister Modi “doesn’t talk about that”, she said.Asked about the constant criticism the Nehru-Gandhi family faces from the BJP, she recounted an anecdote about her guru telling her the calmness that Lord Buddha had in face of detractors.Priyanka was accompanied by her husband Robert Vadra who also cast his vote.The seven constituencies in Delhi are voting on Sunday alongside 14 seats in Uttar Pradesh, 10 in Haryana, eight each in Bihar, Madhya Pradesh and West Bengal, and four in Jharkhand as part of the sixth phase of the 2019 poll which ends on May 19. Counting of vote will take place on May 23.The Congress chief said that so far it has been a “good fight” in the elections.Asked how many seats the Congress is winning, he said, “I won’t comment on that. It will be decided by the people.””People are the master. Whatever people say, we will accept. How many seats we will win, people will decide that… Jai Hind,” he said.In Delhi, it is largely a three-way contest among the BJP, the Congress and the AAP.In the 2014 Lok Sabha poll, all the seven seats in the national capital were won by the BJP. While the Congress is hoping to bounce back in Delhi, the AAP is keen to give a good fight. Bharatiya Janata PartyBJPcongressLok Sabha elections 2019 First Published: May 12, 2019, 1:09 PM IST | Edited by: Ahona Senguptalast_img read more

Amit Shah Should be Made to Squat Holding Ears for Kangal Bangla

first_img Amit ShahBJPCRPFelection commission First Published: May 14, 2019, 9:10 PM IST Kolkata: Trinamool Congress supremo Mamata Banerjee Tuesday said BJP president Amit Shah should be made to squat holding his ears for his ‘Kangal Bangla’ remark, and alleged that his party is transporting money in CRPF vehicles to be distributed among people of the state.She also attacked the Election Commission claiming that efforts to check entry of money into the state were hampered after the poll panel changed the police commissioners of Kolkata and neighbouring Bidhannagar. Wondering whether the BJP president knows the meaning of ‘kangal’, she said, “How dare he say Bengal is kangal…. He should be made to squat holding his ears for saying this,” Banerjee at election rallies here.Shah had on Monday claimed at a BJP rally that under the TMC rule, ‘Sonar (golden) Bangla’ has been turned into “kangal (pauper) Bangla” and that it is the saffron party which can bring back the lost glory of the state.The chief minister did not react to Shah’s ‘kangal Bangla’ comment Monday, but senior TMC leader and Rajya Sabha member Derek O’ Brien hit back calling Shah “puke-worthy” and a “low-life” who has insulted the state.”Money is being transported in CRPF vehicles…. I have never seen the way money is flowing in this election. Such things have been seen only in the North-East but not elsewhere,” Banerjee said at a public meeting.Claiming that crores of rupees brought through hawala, she said it is shameful that money is being used to win votes. Even small leaders of the BJP are getting central police cover and these leaders are using police vehicles to transport money, the TMC supremo alleged.Banerjee claimed that the commissioner of Bidhannagar Police, under whose jurisdiction lies the city airport, was changed to facilitate movement of money.The Election Commission had on April 5 transferred Kolkata Police Commissioner Anuj Sharma and Bidhannagar Police Commissioner Gyanwant Singh ahead of the first phase of polling on April 11. Dr Rajesh Kumar, ADG, Pollution Control Board, was made the new Kolkata top cop.The chief minister had then alleged that it was done at the behest of the opposition BJP leaders. Asserting that the EC will not be there after the polls, she said “I will find then out who have distributed the money and who took those.”She said the new commissioner of Kolkata Police will have to face the consequences if he is found to be partisan. Training her gun on the EC, she claimed that dates for the seven-phase election in the state was fixed as per the wishes of Prime Minister Narendra Modi and BJP president Amit Shah to suit their campaign schedule.She alleged that hooligans from Uttar Pradesh, Bihar and Jharkhand have been brought to West Bengal to foment trouble in the state during the election.”Someone has to bell the cat. I have taken on them and that is why they (BJP) are so angry on us (TMC),” she said. Accusing the BJP leadership of vindictiveness, Banerjee claimed that the saffron party is using the Income Tax authorities and the CBI to scare opposition leaders. last_img read more

Bijayant Panda Likely to Win Kendrapara Says News18IPSOS Survey

first_imgAs the country awaits the results of the world’s biggest democratic exercise, News18-IPSOS, an exit poll survey by Network 18, gives an insight into the probable outcome of the results.News18-IPSOS suggests that BJP candidate Bijayant Panda may emerge as the winner from the Kendrapara seat in the state of Odisha. He is in contest with Anubhav Mohanty who is fielded by BJD. Earlier in 2014, the BJD won this seat when Bijayant Panda defeated Dharanidhar Nayak of Congress with a huge margin of 18.3 per cent votes.The BJD bastion has always voted against the Congress. This was known to be the forte of Biju Patnaik, father of Odisha CM Naveen Patnaik. The entry of BJP into the contest was another twist in the tale of Kendrapara. BJD has always overpowered all the national parties in the state of Odisha.In the previous two election s—2004 and 2009— BJD held the seat in this constituency. In 2004, Archana Nayak of BJD defeated Congress candidate Srikant Kumar Jen with a margin of 10.4 per cent votes. In the 2009 elections, Bijayant Panda of BJD defeated the Congress candidate Ranjit Biswal with a margin of 12.9 per cent votes. The Lok Sabha elections in 2019 took place in seven phases across India in a time span of about a month and a half. A total of 543 seats went to poll in which around 900 million voters were expected to exercise their voting rights.The final results will be announced on May 23. Bijayant PandaBJPcongresskendrapara-s18p15 First Published: May 20, 2019, 7:29 PM ISTlast_img read more

BJP National General Secretary Ram Lal Appointed to RSS Post V Satish

first_img BJPNarendra ModiRam Lalrss First Published: July 13, 2019, 6:56 PM IST New Delhi: Ram Lal, the longest-serving BJP general secretary in charge of organisation, has been repatriated to the Rashtriya Swayamsevak Sangh (RSS). The decision was announced after a three-day meeting of ‘Sangh pracharaks’ at Vijaywada.The post of the general secretary of the organisation, or ‘sangathan mantri’, is the key link between the BJP and the RSS. A full-time RSS ‘pracharak’ is appointed to the post who works on deputation in the BJP till his repatriation. Ram lal will now function as the ‘seh sampark pramukh’ (or co-convenor of all-India public relations). In the past Prime Minister Narendra Modi, Sanjay Joshi, Govindacharya and Kushabhau Tjackray have worked in the post.There were four national joint general secretaries (organisation) under Ramlal — V Satish, Saudan Singh, Shiv Prakash and BL Santhosh. Satish is likely to be announced as his successor.Arun Kumar, Akhil Bharatiya public relations chief, said, “He is no more in the responsibility of the BJP. He will be our all-India executive member looking after public relations. The decision was taken today at the conclusion of the three-day long meeting.”“It is now for the BJP to decide on the next general secretary. We have followed our regular practice. These changes take place after some years, there is no specific reason behind this,” Kumar added.Ram Lal, the longest-serving general secretary, is known for his organisational skills and coordination with the RSS while keeping in mind the primacy of the Sangh ideology. His colleagues laud him for being rooted to the ground and for calling frequent meetings. He was well-connected with the state units and kept the channels of communication open between the BJP and RSS. last_img read more

Facebooks Libra could be dead on arrival if India stands by its

first_imgJust as the initial buzz around Facebook’s newly unveiled Libra cryptocurrency started to die down, it roared back to life this week with India’s announcement that it is considering a ban on the cryptocurrency within its jurisdiction. If the prohibition were to go into effect, Libra would be kept out of the third-largest market in the world’s most populous continent, which would seriously imperil the fledgling digital currency. This adversarial stance by the Indian government also comes only days after the U.S  House of Representatives issued a request for Facebook to delay Libra’s launch so that possible regulatory measures could be devised, debated, and enacted.Considering the segment of consumers Libra targets, India’s proposed ban poses an especially dire threat. In the cryptocurrency’s debut last month, Facebook noted that Libra was designed to allow smartphone users without reliable access to bank accounts to exchange fiat currency into Libra to facilitate online purchases. By giving so-called “unbanked” adults around the world a frictionless way to engage in online commerce, Facebook would ostensibly benefit from essentially dominating the market for such a solution.When weighed against Libra’s raison d’etre, India occupies the sweet spot between low bank account use, exploding smartphone adoption, and a large and accelerating economy. In fact, India’s market makes for such a perfect fit for Libra that it is a near certainty that Facebook had the country in mind to drive phase one of the rollout. Without India, Libra’s hopes could be dashed before a single coin is traded.It might seem puzzling at first blush to see India taking such a hard line on Libra, but it makes more sense when viewed in the context of Facebook’s dubious history expanding in the developing world and India’s bid for global economic supremacy. As Facebook’s uptake in the developed world has basically hit its peak, the social network has been desperately trying to dig out a foothold in developing regions.Amy Osborne/Getty Images)These initiatives have usually been cloaked in the guise of altruism. Upon further inspection, they tend to be near exploitative beneath the surface. The most notable instance of this was Facebook’s developing world internet connectivity project, facebook.org, which was found to only allow access to only a handful of websites with much of their encryption stripped out (ostensibly to open user data to collection and mining). Any country would understandably be loath let a company that shows such contempt for users’ privacy and security do business within its borders.On top of that, India is vying to become an economic superpower, and allowing a foreign-controlled digital currency to set up shop in the country would be seriously at odds with those plans. Part of growing an economy to superpower proportions involves strengthening domestic banking institutions, particularly to the point where they extend more and more credit abroad. The last thing India, or any nation in a similar position, would want is to stifle the growth of its domestic banking sector. It’d be doing exactly that by opening it to competition with a banking alternative backed by such foreign payment processing heavyweights as Visa, MasterCard, and PayPal.Until India receives assurances from Facebook and the Libra Association that Libra won’t abuse its citizens’ digital privacy, and won’t harm its domestic banking interests (the latter guarantee being much harder to ensure than the former), India seems unlikely to budge on the its promised ban.The views expressed here are solely those of the author and do not reflect the beliefs of Digital Trends. Editors’ Recommendations What is Libra? Here’s what you need to know about Facebook’s new cryptocurrency Congress already wants to block rollout of Facebook’s cryptocurrency Congressional Democrats demand Facebook halt its Libra cryptocurrency rollout Facebook says it won’t launch Libra until regulators are happy President Trump attacks Facebook Libra, says it’s not dependable like the dollarlast_img read more

Amazon workers AIpowered cat flap stops his pet from bringing in dead

first_imgHow do you stop your beloved pet cat catching and murdering innocent wildlife, then bringing it into your home in the middle of the night and waking you up? There’s plenty of advice out there, ranging from getting sweet little Figaro a collar with a bell, to disincentivizing the act by quickly disposing of dead critters without making a fuss. Of course, if you’re a technically minded Amazon employee, there’s another option too: You rig up your cat flap to an A.I.-enabled camera and Arduino-powered smart locking system. You know, the good, old fashioned way.That’s exactly what Amazon product manager Ben Hamm did recently, after reaching the end of his tether with cute-yet-murderous cat Metric. To create the door locking mechanism, Hamm affixed an Arduino to an electromagnetic bolt that he attached to the cat door. He also set up a camera using Amazon’s AWS DeepLens, a programmable video camera which can run deep learning models locally and take action according to what it sees. He then set about coding his solution using programming language Python and Amazon SageMaker, a managed cloud service that enables developers to build, train, and deploy machine learning models.“The most important part of training machine learning software is getting your hands on enough data,” Hamm told Digital Trends. “First I programmed my camera just to capture images every time it detected movement, and then I spent months labeling over 23,000 images. As the most important data — Metric bringing in prey — happened unpredictably, this took time. If my camera failed when Metric brought in a rat, I would get disappointed. You start to get very greedy for quality data! Once I had the data nice and labeled, I fed it into Amazon SageMaker to create image classification models.”The finished system means that, when Hamm’s camera detects Metric trying to bring in prey, it sends a USB signal that locks the door. (Because DeepLens does its processing locally, this will work even if Hamm’s internet goes down.) The cat flap remains locked for a quarter hour, and Hamm is sent a text message. He also set it up so that it sends a “blood money” donation to the National Audubon Society, which protects birds.As to whether he plans to commercialize the technology, Hamm remains unsure. “The finished product is currently installed over the window of my cottage,” he said. “I was not planning on commercializing it, but the outpouring of desperate cat owners has definitely been interesting!” Editors’ Recommendations The iKuddle is a smart litter box that scoops the poop for you Cringeworthy Cats trailer reminds us we’re not out of the Uncanny Valley yet Protect your house with the best home security cameras of 2019 A fond farewell to Grumpy Cat, the internet’s most famous feline Eight Sleep Pod Review: This smart bed is designed for coupleslast_img read more