Daily Archives: March 9, 2017

Machine learning is the next big thing, says Google with its acquisition of AI site Kaggle.

It doesn't take artificial intelligence to know Google thinks machine learning will be central to your future.

After all, the Silicon Valley powerhouse has been busy creating self-teaching tech that can translate languages, vamp with you on piano, and politely crush you at the ancient Chinese game of Go.

"Over time, the computer itself -- whatever its form factor -- will be an intelligent assistant helping you through your day," Google CEO Sundar Pichai wrote in his first-ever letter to shareholders, last year. "We will move from mobile first to an AI first world."

Now Google has taken another step toward that future. On Wednesday, the Google Cloud Platform said it had acquired Kaggle, what it calls the world's biggest community for data scientists and machine learning geeks.

Among other things, Kaggle lets AI enthusiasts "climb the world's most elite machine learning leaderboards," "explore and analyze a collection of high quality datasets," and "run code in the cloud and receive community feedback on your work," according to the site.

The Kaggle team will stay together and continue Kaggle as its own brand within Google Cloud, Kaggle CEO Anthony Goldbloom said in a blog post.

Fei-Fei Li, chief scientist, Google Cloud AI and machine learning, said in her own post that the acquisition would give Kaggle members direct access to the most advanced cloud machine learning environment.

"We must lower the barriers of entry to AI and make it available to the largest community of developers, users and enterprises, so they can apply it to their own unique needs," Li wrote. "With Kaggle joining the Google Cloud team, we can accelerate this mission."

So much for avenging your Go loss. Tennis, anyone?

Read more from the original source:

Google buys Kaggle and its gaggle of AI geeks - CNET

Top generals from Russia, Turkey, and the U.S. met yesterday in Turkey to discuss tension over the partnerships with local forces fighting ISIS in Syria. Late last week, a buffer zone just outside of Manbij, Syria was created that separates U.S.-backed Kurdish forces, the YPG, from clashes with Turks.

Turkey considers the YPG a terrorist group and said last week it would capture Manbij in a "confrontation" if the U.S. didn't kick out its local fighters.

Why this matters: Instead of splitting their attention between fighting Turkey and ISIS, the U.S.-backed forces can now fully focus on ISIS. Plus, this is a Russian-brokered deal, which is worth noting since this is not the last time rival rebel groups fighting ISIS will clash with one another as they all move closer in on ISIS geographically.

The U.S. called the agreement "a measure of success." Turkey? Not so much: "What Turkey is experiencing with its allies in the West is traumatic We do hope the Trump administration will have a better understanding of Turkey's concerns," Turkey's Deputy Prime Minister Mehmet Simsek said.

Visit link:

The head of Uber's AI labs is latest to leave the company - Axios

Facebook today announced a new server design it calls Big Basin, a successor to its Big Sur line of artificial intelligence training systems. These Nvidia-powered GPU servers, tied together into large training networks for AI software, are what enable Facebook products to perform object and facial recognition and real-time text translation, as well as describe and understand the contents of photos and videos.

Big Basin can now train on learning models 30 percent larger than its predecessor, Facebook says. It can also crunch through the massive number sets used by an AI system to improve itself at nearly twice the speed, according to tests conducted on standardized neural network models.

Facebooks AI training systems just became faster and more capable

Facebook plans to make the server design open to the public in the near future. Thats standard at the company, which participates in and helped create the Open Compute Project for sharing and collaborating on data center hardware and software. So anyone even server design specialists in competing companies will soon be able to download the Big Basin schematics once theyre posted online.

For Facebook, its less about keeping under wraps the tools it uses to train AI systems and more about trying to advance what its AI systems are capable of. Its not just about pushing the limits of technology, though Facebook is among one of the largest organizations investing in cutting-edge and experimental AI research. The companys large investments in AI go hand-in-hand with its push toward live video and other consumer-centric focuses. If youve logged into Facebook, its very likely youve used some type of AI system weve been developing says Kevin Lee, a technical program manager at Facebook who works on Big Basin and other data center initiatives.

For instance, by tagging friends and categorizing videos including those streamed live Facebook may help drive more users to upload video and consume it. Theres also a large social impact the company can have with its AI research. One key function of Facebooks current AI algorithms today is describing the contents of photos to blind users, while just last week Facebook announced it would use AI-powered pattern recognition software to try and identify when troubled users may be in need of mental health outreach. All of this is made possible because the company invests in and continues to develop the servers, like Big Basin, that train these systems before theyre pushed out to public products.

View post:

Facebook's new AI training server is nearly twice as fast - The Verge

Chinas Uber rival Didi Chuxing has officially opened its U.S.-based research lab. The new center is part of a move to suck up talent beyond Didis current catchment pool in China, particularly in the areas of AI and self-driving vehicles, but it doesnt signal an expansion of itsservice into North America.

The existence of the research center itself isnt new. Last September, TechCrunch wrote that Didi had hired a pair of experienced security experts based in the U.S. Dr Fengmin Gong and Zheng Bu to lead the center, which works closely with another China-based facilitythat opened in late 2015, but now it is officially open.

Dr Gong will lead the facility in Mountain View, and his team of dozens of leading data scientists and researchers will include former Uber researcher Charlie Miller. Miller rose to fame in 2015 when he hacked a journalists vehicle from a laptop 10 miles awayin a pre-arranged stunt to demonstratevulnerabilities within the automotive industry.

Millers job seems much like his role at Uber according to tweets he sent out today. His defection is noteworthy since it appears to be the first major poach that Didi has made from Uber, and it falls in the self-driving car space whereUber has made a huge push.

Didi is looking to make an early impact in Silicon Valley through a partnership with Udacity around self-driving vehicles. The two companies announced a joint contest inviting teams to developan Automated Safety and Awareness Processing Stack (ASAPS) to increasedriving safety for both manual and self-driving vehicles. Five finalists chosen will get a shot at the $100,000 grand prize and the opportunity to work closer with Didi and Udacity on automotive projects.

Read this article:

China's Didi Chuxing opens US lab to develop AI and self-driving car tech - TechCrunch

Photo: Tekla Perry Teal's drone uses the Nvidia Jetson module to identify what its cameras see

People arejust now getting comfortable with the ideathat data from many electronic gadgets they use flies up to the cloud. But going forward, much of that data will stick closer to Earth, processed in hardware that lives at the so-called edgefor example, inside security cameras or drones.

Thats why Nvidia, the processor company whose graphics processing units (GPUs) are powering much of the boom in deep learning, is now focused on the edge. Deepu Talla, vice president and general manager of the companys Tegra businessunit, says bringing AI technology to the edge will make a new class of intelligent machines possible. These devices will enable intelligent video analytics that keep our cities smarter and safer, new kinds of robots that optimize manufacturing, and new collaboration that makes long-distance work more efficient, he said in a statement.

Why the move to the edge? At a press event held Tuesday in San Francisco, Talla gave four main reasons: bandwidth, latency, privacy, and availability. Bandwidth is becoming an issue for cloud processing, he indicated, particularly for video, because cameras in video applications such aspublic safety are moving to 4K resolution and increasing in numbers. By 2020, there will be 1 billion cameras in the world doing public safety and streaming data, he said. Theres not enough upstream bandwidth available to send all this to the cloud. So, processing at the edge will be an absolute necessity.

Latency, he said, becomes an issue in robotics and self-driving cars, applications in which decisions have to be made with lightning speed. Privacy, of course, is easier to protect when data isnt moving around. And availability of the cloud, Talla pointed out, is an issue in many parts of the world where communications are limited.

We will see AI transferring to the edge, he said, with future intelligent applications using a combination of edge and cloud processing.

Nvidia, of course, wasnt painting this glowing picture of edge computing without some self-interest. At the event, the company announced its new edge-processing platform, the Nvidia Jetson TX2. This credit cardsizemodule is a plug-in replacement for the companys Jetson TX1, designed for embedded computing. Depending on how it is applied, it can either run at twice the speed of its predecessor or use half the power. Detailed specs are here. The developer kit costs $600, $300 for educators; the production version will sell for $400.

Developers, showing off their work at the launch event, were happy to point out how they are using or intend to use internal AI processing at the edge. A few examples:

IEEE Spectrums blog featuring the people, places, and passions of the world of technologists in Silicon Valley and its environs. Contact us:t.perry@ieee.org

University of Michigan "micromotes" aim to make the Internet of Things smarter without consuming more power 10Feb

Inventors of the CCD, the pinned photodiode, and the CMOS imager honored with 1 million prize 2Feb

Ubers experiment in San Francisco showed that bicycles and bike lanes are a problem self-driving cars are struggling to crack 31Jan

Nvidia CEO Jen-Hsun Huang wants his robocar computer package to become an industry standard 5Jan

Supercharged hardware will speed up deep learning in everything from tiny devices to massivedatacenters 29Dec2016

Microchips AWS-ECC508 bakes in securecommunications 22Dec2016

Giving each pixel hundreds of memory cells means CMOS imaging chips will have much faster global shutter speed 16Dec2016

Danish island experiments with demand aggregation 7Dec2016

The company will expand the autonomous-taxi experiment it began some months ago in Singapore 21Nov2016

Seeing isnt just about taking pictures. The real revolution will come when our digital devices understand whats in front of their eyes 3Nov2016

The "Jeep hacker" says denial-of-service attacks against cars are easy hacksand urges people not to buy any car dongles 28Oct2016

AI-powered traffic light coordination would cut time spent in the car, traffic congestion, and emissions 17Oct2016

Self-driving cars require driver-monitoring capability to know when it is safe to hand over control 4Oct2016

A revised draft of regulations proposes doing away with drivers during tests 3Oct2016

If your smart glasses contain a virtual companion who knows exactly what you are doing and has an opinion about it, are you living in an augmented reality? 30Sep2016

Hondas new Acura NSX is the first to marry a V-6 engine to three electric motors for high-speed steering 26Sep2016

A squishy underwater robot with limbs that bend in every direction requires unusual control strategies 15Aug2016

Industry leaders from around the globe use multiphysics simulation to stay ahead of the curve

Applications streamline capacitive touchscreen design and optimize the workflow

Hear from simulation experts about their experience with computational apps and how they are creating the future of numerical simulation

Here is the original post:

Nvidia wants AI to Get Out of the Cloud and Into a Camera, Drone, or Other Gadget Near You - IEEE Spectrum

Its an insult to people named Trish everywhere.

Even in its success among Silicon Valley nerds, Soylent, the Los Angeles-based meal-replacement drink company, has become something of a parody. In staying true to form the company has invented Trish, its new AI-powered personal assistant. Its mostly useless, but it may answer some of the queries you may have about Soylent products.

No need to buckle your seatbelt, though, the questions you can ask are pretty basic. And the answers are mostly arthritic.

According to the gospel of Soylent, maximum nutrition should be delivered with minimal effort. And while that might be a noble ambition among some crowds, its also led to a fair number of embarrassing headlines, including that time Soylent bars gave people explosive diarrhea. Still, Soylentnamed after the processed food in the sci-fi film Soylent Green that turns out to be made from humanspersists.

The new botcreated by the marketing firm Wieden+Kennedywont just be appearing on Soylents website, according to SocialTech. It will also appear on broadcast and social media. Soylents chief marketing officer, Adam Grablick, told AdAge that Trish is part of an effort to bring the brand to life.

Theres an AI bot frenzy happening in companies, especially at the tech firms where Soylent found many of its earliest and most evangelical users. Facebooks Messenger app has a bunch of them and its founder built his own, which he named Jarvis. US telecom AT&T has a new AI-powered bot named Atticus. Coca-Cola wants an AI bot to create its ads, one has been built to beat humans at Texas Hold Em poker, and theres even one out there watching and interpreting porn. Quartz even has one, which we taught to play old-school Atari games.

Still, Trish isnt all that interesting (you can only ask it so many questions), and only time will tell if it becomes useful over time. In the meantime, maybe buy any real-life friends named Trish a stiffer drink than Soylent.

Read the original post:

Just like all the cool kids, Soylent now has its own pointless AI assistant - Quartz