We applied machine learning to analyze 4,607,160 data points collected by scraping the front page of Reddit for 22 days.Read post for insights on how the Reddit algorithms work.Why people are so damn interested in getting to the front page of Reddit?Look at any viral videos floating around Facebook.For example, a Reddit front page giraffe story got on CNN.com homepage in 2 hours.Our Reddit Scraping Process to get 4 million data points :
Using a special 3D camera, a Tango equipped device looks at the world around it, and uses that information to map a space, add in augmented reality objects, or simply track the location of a mobile device.By building a memory of the world around it, the Tango device can recognize those spaces, and properly fit objects back into them.Several demos served to illustrate how well this worked.Turning on area learning produced entirely different results.Granted, it sometimes took a few seconds to re-identify the space around it, but once it did, the box popped right back up.Memory makes for better multiplayerArea learning really opens up the possibilities of VR and AR gaming with multiple users.
As FindFace has shown, image recognition has made incredible strides over the past few years.Neural networks are brain-like computer systems designed to pick out and create patterns based on inputs.These less-than-accurate representations, be it on purpose or due to lack of drawing skills, often leads to dramatized proportions and missing elements that might not prevent humans from recognizing the image, but causes problems for computers.Using more than 600 individuals, scientists displayed a random image for only two seconds and asked the participant to sketch the pictured object from memory.In the initial testing phase, the program was able to match the doodles up to their original image 37 percent of the time.From a more commercial standpoint, one of the most useful features of this technology would be to help police identify criminals by comparing sketch artists renderings to mugshot databases.
Google Classroom is being joined by a new Coursework API that will allow developers to have their applications further integrate with the aforementioned Classroom service.Classroom is Google s educational management system for teachers and schools, and it uses the company s own Drive platform.Today, Google has announced updates to its Classroom offering, saying that, among other things, reporting systems and gradebooks can now be setup to sync students grades with Google Classroom, removing the time sink of manually transferring grades.Developers also have the newly gained ability to store files in the same Drive folder as the rest of the resources in a class, or use course groups to manage file sharing permissions, says Google.The company already has some partners in testing with the new features, including GeoGebra, OpenEd, and Tynker, all of them being educational platforms.Until then, check out our Google I/O portal for all the news fresh off the ground floor.
The breakthrough involves phase-change memory PCM which IBM has successfully achieved storing three bits of data per cell for the first time, compared to previous demonstrations of storing one bit per cell.PCM is much more durable than flash – it can last something like 10 million write cycles compared to the average flash USB stick which endures around 3,000 write cycles – and it's way faster, coming closer to DRAM performance, but with one big difference: it doesn't lose data when switched off like DRAM.Significant cost reductionDr Haris Pozidis, manager of non-volatile memory research at IBM Research, commented: "Phase change memory is the first instantiation of a universal memory with properties of both DRAM and flash, thus answering one of the grand challenges of our industry.The new memory tech could have implications across a range of uses, including providing blazingly quick storage for cloud and IoT applications, and boosting performance of the likes of machine learning.Businesses could see entire databases stored in PCM enabling ultra-fast querying, and of course this will also make a major difference to smartphones.IBM envisions hybrid applications with PCM running alongside traditional flash storage in a phone, but with the OS stored in the PCM so when you switch your phone on, it loads almost immediately and you're staring at your home screen before you've had time to bat an eyelid.
Experience can only take you so far — the dynamic nature of the world we live in has made specialized training a must.If you re a marketer who wants to stay competitive, you ve got to keep your digital marketing skills on the cutting edge.What you get with Simplilearn s Full Stack Masters ProgramThe Full-Stack Digital Marketer Masters Program includes:access to the foundational Digital Marketing Certified Associate course and project work;nine advanced courses including PPC, web analytics, SEO, content and email;monthly mentoring sessions by experts;on-the-go learning with Android and iOS apps; andMasters Program certificate from Simplilearn.Simplilearn s Digital Marketing Certified Associate course – previewMarket Motive s Digital Marketing Expert PassFounded in 2007, Market Motive is a pioneer in online digital marketing training.But the course s open availability, lack of mentored guidance and access to expert instructors make completing the specialization challenging for self-learners.Choose the best provider for your needsWhether you just need a quick beginner s primer or are ready for a deep dive into advanced digital marketing topics, you are sure to find something to match your learning style and your needs.
Credit: Derek WalterIf you re not one of the lucky few who gets to attend Google I/O this week, don't despair.The keynote will be streamed live, and we were kind enough to embed the video stream below.If you want to really get crazy, you can slap on a pair of Cardboard goggles as the event will be shot in 360-degree video.Many of the other events and developer sessions will be live streamed from the official I/O site.Sessions include What s New in Android, Introducing Project Tango Area Learning, and Android Auto: The Road Ahead.Better yet, Greenbot is on the ground here at the Shoreline Amphitheatre, so follow us here and on Twitter to get live updates throughout Google I/O.
GIFMastering the game of chess is infinitely complicated, but getting started just got easier with a chess board that s perfectly set up and ready to go as soon as you open the box.Before newcomers can even begin to start learning how the various chess pieces move around the board, they have to know what squares the pawns, bishops, rooks, knights, kings, and queens all call home.GIFAs a game of chess plays out, captured pieces are returned to their designated and labeled slots in the game s box.The bottom of each piece, and the double-sided chess board, are all magnetic.The game s design also helps ensure that every piece is accounted for at the end of a game, and that none of them go missing when it s being haphazardly stuffed into your game shelf.It s not indestructible, but should certainly survive being swept off the table in a rage every time your partner puts you in check.
We have had good success with our GPUs in high-performance computing, deep learning in hpc, deep learning, data analytics, remote work stations, said McHugh, a former Cisco executive who joined Nvidia six months ago.The Tesla M10 GPU has high user density when it comes to delivering apps such as Outlook, Office 2016, web browsers, Adobe Photoshop, and the Windows 10 operating system.Delivering business applications in a virtualized way is becoming more challenging because more businesses are using demanding graphics apps these days.The percentage of GPU-accelerated apps has more than doubled in the past five years, with half that growth coming in the first months of 2016 alone, according to a study by Lakeside Software.To provide the best user experience, these applications increasingly use OpenGL and DirectX APIs, as well as graphics technology from the data center.While the need for advanced GPU technology has commonly been associated with the usage of 3D applications, as enterprises make the move to software like Windows 10, Office 365, and other SaaS and web apps, IT departments will increasingly seek the benefits of GPU acceleration to provide everyday business tools to all of their users, said Robert Young, analyst for IT Service Management and Client Virtualization Software at IDC, in a statement.Nvidia is teaming  up with virtualization software companies, such as Citrix and VMware, to deliver a high-end virtualized app that runs as if it were being processed on a user s personal machine.The cost of running such virtual apps or remote desktop sessions is now down to less than $2 a month per user and, for virtual PCs, is less than $6 a month per user.The new Nvidia Grid software is available worldwide today, and the Tesla M10 will be generally available in the fall.Virtualized apps can now be delivered at a subscription price of about $10 per concurrent user, McHugh said, on the Nvidia Grid service.
The instruments we used to test these machines are capable of detecting particles as small as 0.010 micron, which is far below the threshold for civilian equipment and below the 0.3 micron threshold for the HEPA standard.What follows is just a summary of our findings, but for those who are interested in learning more, the How we tested section of our full review includes our procedures, methods, and results, complete with tables and graphs of our results.The thin 3/16-inch activated carbon filter is just not up to the task.If you have severe allergies to particulate allergens or other health issues related to air quality, or you re willing to pay a premium for a machine that can be extremely efficient while running quietly, we recommend our upgrade pick, the Blueair 503 with the SmokeStop filter package.In our noise-limited tests, it removed more than 90 percent of the particles from our bedroom-sized testing area in 10 minutes—that s over 10 percent more than the next best purifier and over 60 percent more effective than the control.Its exceptional performance in this area is a big part of why FEMA and the Red Cross chose Austin Air units for deployment at Ground Zero and the surrounding areas in the aftermath of 9/11.
At its Google I/O developer conference in Mountain View, California, today, Google is announcing a new tool for developers to use if they d like their applications to integrate with Google Classroom, a cloud service for handling assignments, coursework, and grades.Learning tools can focus on creating great content, and use Classroom to manage the workflow for assignments created with their content.And the existing Classroom API is getting more powerful — it will be able to work with the Google Drive folder for a given course.Applications can use this new functionality to store files in the same Drive folder as the rest of the resources in a class, or use course groups to manage file sharing permissions, Kupershlak wrote.The enhancements come as other companies strive to make their core products more widely relied upon in the education sector.In March Apple launched a Classroom app for iOS alongside the release of iOS 9.3, and in April Microsoft launched an app called Classroom in preview for Office 365 Education customers.Earlier this month Dropbox, which competes with Google Drive, launched an Education tier.Google has gained share with its Chromebooks, which lean heavily on cloud storage that companies provision through Google Apps for Education.
Google has begun to build its own custom application-specific integrated circuit ASIC chip called tensor processing units TPUs , Google chief executive Sundar Pichai said today at the Google I/O developer conference in Mountain View, California.The name is inspired by Google s TensorFlow open source deep learning framework.When you use the Google Cloud Platform, you can take advantage of TPUs as well, Pichai said.Specialty hardware — sort of taking a cue from the holographic processing unit HPU inside Microsoft s HoloLens augmented reality headset — will not be the only thing that will make the Google public cloud stand out from market leader Amazon Web Services AWS .Also, over time Google will expose more and more machine learning APIs, Pichai said.Our goal is to lead the industry on machine learning and make that innovation available to our customers, Google distinguished hardware engineer Norm Jouppi wrote in a blog post.Building TPUs into our infrastructure stack will allow us to bring the power of Google to developers across software like TensorFlow and Cloud Machine Learning with advanced acceleration capabilities.Machine Learning is transforming how developers build intelligent applications that benefit customers and consumers, and we re excited to see the possibilities come to life.
Hello from Google I/O, Google s annual developers conference.This year, Google I/O has moved from its traditional home in San Francisco to an outdoor setting: the Shoreline Amphitheater in Mountain View, California — across the street from Google s main Googleplex headquarters.It s expected official details of Google Home will be announced, Google s voice activated rival to the Amazon Echo.Machine learning will likely come up, including Google perhaps finally getting into the bot game.You can watch the keynote through the live stream at the Google I/O site or follow along as we live blog it from the event itself, when it begins at 10am PT.Our live blog is below to make the font larger, click the gear icon on the right :
Fitness apps will start automatically, and users can respond to messages with machine learning-based repliesAndroid Watches displayed on screen at Google I/O on May 18, 2016Android Wear, Google s operating system for smartwatches, is getting its biggest update yet with an upcoming 2.0 release that brings improved features for messaging and fitness.Among the improvements, Android Wear 2.0 will detect when you're starting to exercise and automatically fire up an app such as Strava.Google is also taking aim at how the Apple Watch displays information from applications, by allowing developers to display information from any app when users glance at their device.That s important for helping people to quickly see data from apps they use frequently.People who want to converse with friends from their wrists will also get new features, including a redesigned keyboard, support for handwriting recognition, and smart replies that offer machine learning-driven responses based on the context of a conversation.At a time when smartwatches are still a niche item, that could be useful for Google.
The Lego-like ArcaBoard, which went into production last month, can float about 20 inches above the ground.Too bad it costs $20,000 and offers a range of about a mile.The answer: nothing nearly as practical as cruising around Hill Valley or fleeing a gang of bullies.Four turbojets provide a blustery 1,000 horsepower, enough thrust to keep a rider suspended and zipping along.What It s ForWell, if riding a bike required strapping jet fuel to your back, risking broken bones with every spill, and dedicating a good week to learning how to just get down the street.To catch a glimpse at the device s more immediate future, it s instructive to look at another product from Zapata s company, the Flyboard Sport.
AmazonWhile still a primarily online store, Amazon isn't turning its back on physical retail establishments.After opening its first brick-and-mortar store in Seattle last year, the company's CEO Jeff Bezos confirmed at Amazon's shareholder meeting that more stores are coming."We re definitely going to open additional stores; how many we don t know yet, Bezos said at the meeting according to a report from The Wall Street Journal.In these early days, it s all about learning rather than trying to earn a lot of revenue."Currently Amazon's Seattle location is mostly a bookstore, and the company is already building another location in San Diego.Still, Bezos' statement is a bold one, so we'll likely see numerous additional Prime membership perks rolling out soon.
You're probably giving away more than you thinkThe location stamps on just a handful of Twitter posts can help even low-tech stalkers find you, researchers found.The notion of online privacy has been greatly diminished in recent years, and just this week two new studies confirm what to many minds is already a dismal picture.First, a study reported on Monday by Stanford University found that smartphone metadata -- information about calls and text messages, such as time and length -- can reveal a surprising amount of personal detail.Based on frequent calls to a local firearms dealer that prominently advertises AR semiautomatic rifles and to the customer support hotline of a major manufacturer that produces them, it's logical to conclude that another likely owns such a weapon.Currently, U.S. law gives more privacy protections to call content and makes it easier for government agencies to obtain metadata, in part because policymakers assume that it shouldn t be possible to infer specific sensitive details about people based on metadata alone.Many people have this idea that only machine-learning techniques can discover interesting patterns in location data, and they feel secure that not everyone has the technical knowledge to do that, said Ilaria Liccardi, a research scientist at MIT s Internet Policy Research Initiative and first author on the paper.
The location stamps on just a handful of Twitter posts can help even low-tech stalkers find you, researchers found.The notion of online privacy has been greatly diminished in recent years, and just this week two new studies confirm what to many minds is already a dismal picture.First, a study reported on Monday by Stanford University found that smartphone metadata—information about calls and text messages, such as time and length—can reveal a surprising amount of personal detail.Based on frequent calls to a local firearms dealer that prominently advertises AR semiautomatic rifles and to the customer support hotline of a major manufacturer that produces them, it s logical to conclude that another likely owns such a weapon.Currently, U.S. law gives more privacy protections to call content and makes it easier for government agencies to obtain metadata, in part because policymakers assume that it shouldn t be possible to infer specific sensitive details about people based on metadata alone.Many people have this idea that only machine-learning techniques can discover interesting patterns in location data, and they feel secure that not everyone has the technical knowledge to do that, said Ilaria Liccardi, a research scientist at MIT s Internet Policy Research Initiative and first author on the paper.
Credit: GoogleForget the CPU, GPU, and FPGA, Google says its Tensor Processing Unit, or TPU, advances machine learning capability by a factor of three generations.This is roughly equivalent to fast-forwarding technology about seven years into the future three generations of Moore s Law , the blog said.Because of this, we can squeeze more operations per second into the silicon, use more sophisticated and powerful machine learning models, and apply these models more quickly, so users get more intelligent results more rapidly.The tiny TPU can fit into a hard drive slot within the data center rack and has already been powering RankBrain and Street View, the blog said.Analyst Patrick Moorhead of Moore Insights & Strategy, who attended the I/O developer conference, said, from what little Google has revealed about the TPU, he doesn t think the company is about to abandon traditional CPUs and GPUs just yet.He likened the comparison to decoding an H.265 video stream with a CPU versus an ASIC built for that task.
Answer by Scott Aaronson, Theoretical computer scientist at MIT, soon to be at UT Austin, on Quora.There are things like Deflategate or manspreading or the dresses worn at the Oscars, which many people talk about but few should.And then there are things like World War II, global warming, black holes, or machine learning, which many people talk about and probably many should.Indeed, I started out in AI and machine learning, as an undergrad at Cornell with Bart Selman and then as a grad student at Berkeley with Mike Jordan, before shifting into quantum computing, where I felt like my comparative advantage was greater.On the other hand, at least according to the ML researchers I know, the recent progress has not involved any major new conceptual breakthroughs: it s been more about further refinement of algorithms that already existed in the 70s and 80s, and of course, implementing those algorithms on orders-of-magnitude faster computers and training them with orders-of-magnitude more data.In the end, I suppose it s less interesting to me to look at the sheer amount of machine learning hype than at its content.
More

Top