Friday, 2 June 2017

How hacked computer code allegedly helped a biker gang steal 150 Jeeps

In a cross-border auto heist that resembles a scrapped plot from the “Fast and the Furious” franchise, nine members of a Tijuana-based biker club have been charged with stealing 150 Jeep Wranglers using stolen computer code and key designs, the Justice Department announced earlier this week.
Known as the Hooligans, the biker gang allegedly stole the Jeeps in the San Diego area over the past several years, selling the vehicles or stripping them for parts across the border in Mexico, U.S. Attorney Mark Conover said during a news conference recorded by the San Diego Union-Tribune. The value of the stolen Jeeps was $4.5 million.
According to the indictment, the Hooligans staked out vehicles days before the thefts to obtain their vehicle identification numbers. With these numbers in hand, the suspects were able to get details to create duplicate car keys, as well as the codes needed to program the keys, linking them to the Jeep Wranglers. The key designs and codes were stored in a proprietary database. But law enforcement officials don’t know how the Hooligans were able to access it.
In the course of the investigation, authorities said they learned that nearly 20 requests for duplicate keys were made by a Jeep dealership in Cabo San Lucas, Mexico.
Conover said the thefts took only minutes. After using the duplicate key to get inside the car, the Hooligan members used a handheld electronic device to pair the key with the car's computer to turn the engine on and drive off.
While Conover did not name the exact device used in the thefts, Kathleen Fisher, a Tufts University computer science professor and security researcher, said that such key programmers are relatively cheap, with some costing less than $100, and readily available online.
That auto companies or their partners maintain databases to store key and programming codes is not in itself unusual. After all, rightful car owners would need that information to create new keys if they were locked out, Fisher said. But in this case, it appears the security vulnerability may have been the integrity of the database. One way for criminals to extract stored information is to hack into a network that has access to it, she said. Another way is to get authorized users to obtain the information themselves and then pass it on, or to share active credentials with someone who shouldn't have them.
Experts say that widespread hacks of cars may soon become a reality. In an alarming demonstration captured by a widely read Wired article from 2015, researchers Charlie Miller and Chris Valasek showed that they could wirelessly hijack a 2014 Jeep Cherokee. The researchers could disengage the Jeep's brakes, cause the transmission to malfunction and, at lower speeds, kill the engine altogether.
Hacking tools are easily spread online, and pervasive software threats are costly to patch up. Car companies also face the challenge of justifying increased security costs to customers, Fisher said. A car's cybersecurity isn't the easiest thing to advertise, compared to say, horsepower or leg room. Outside of industry-wide pressure from regulators or insurers, individual companies may hesitate to spend more on security, despite the massive risks that hijacked and hacked cars pose. “We don’t do a very good job accounting for the cost of bad security," Fisher said.

You Don’t Have to Major in Computer Science to Do It as a Career

Basic economics suggests that if college students see booming demand for specific skills, a stampede to major in such lucrative fields should ensue. For years, tech companies, banks, and even traditional industrial companies have been hiring programmers, software developers, and computer scientists as fast as they can find them. Since 2010, there has been a 59 percent leap in jobs for software application developers—and a 15 percent jump in pay, to an average $102,300 last year—according to the U.S. Bureau of Labor Statistics. Accounts of tech engineers earning more than pro athletes keep making headlines.
So why aren’t more U.S. college students majoring in computer science?
U.S. colleges and universities graduated only 59,581 majors in computer and information sciences in 2015, the most recent year for which data is available, according to the National Center for Education Statistics. While that tally grew 7.8 percent from the year earlier, from employers’ reports it does not seem to be keeping up with demand.
Attempts to explain what looks like a chronic training deficit are plentiful. Theories touch on everything from worries that the computer-science curriculum is too hard to apprehension about gender bias in the field. But an extensive new study indicates that both students and employers are finding a way around the problem: making brisk use of less obvious career pathways that lead to software jobs anyway.
Percentage of Graduates Working as Software Developers by Undergraduate Major
  • 5.6% Aerospace Engineers
  • 8.1% Astronomy & Astrophysics
  • 30.3% Computer Engineering
  • 11.3% Electrical Engineering
  • 6.1% Mathematics
  • 8.2% Physics

  • The study, published in May by the Brookings Institution’s Hamilton Project, used U.S. Census Bureau data to track the career choices of 1.2 million college graduates, as observed from 2010 to 2013. Among its findings: many people working as computer scientists, software developers, and programmers used their college years not to major in computer programming or software development, but instead to major in traditional sciences or other types of engineering.
    Among graduates with degrees in physics, math, statistics, or electrical engineering, as many as 20 percent now work in computing-based fields. At least 10 percent of people who majored in aerospace engineering, astronomy, biomedical engineering, or general engineering have made the same migration.
    Even geography, nuclear engineering, and chemistry departments send 3 to 5 percent of their undergraduate majors into software development or similar fields, the Hamilton Project reports.
    At Indiana University Bloomington, dozens of math and science majors have been winning software-sector jobs after graduation, reports Joseph Lovejoy, head of the school’s Walter Center for Career Achievement. Bioinformatics companies such as Cerner and Epic Systems have been keen to hire biology majors who picked up coding skills without majoring in computer science, he adds. General Motors has been recruiting math majors for jobs as software testers and software developers.
    Math majors are in demand at Microsoft too. Dawn Klinghoffer, who tracks hiring trends for the giant software company, explains that fast-growing areas such as machine learning hinge on the ability to create and fine-tune highly sophisticated algorithms. That’s increased Microsoft’s willingness to consider candidates who learned programming on their own but have a deep mastery of complex math.
    More broadly, Klinghoffer says, Microsoft has been “expanding the pool” in its recruiting to help build talent without constantly being caught up in bidding wars against other tech giants trying to hire the same computer-science majors from the same few elite schools. Widening the range of majors also helps create a workforce with more diverse perspectives, Klinghoffer says.
    Among the people taking an unusual path is Luke Kanies, who majored in chemistry at Reed College. Unsure what he wanted to do after college, he managed a series of corporate data centers for about five years, before founding Puppet Labs, a software-management company that helps big companies keep hundreds of overlapping programs as up to date and compatible as possible.
    Kanies portrays his unorthodox beginnings as an asset. At Puppet Labs, he and colleagues test software the same way chemists test their theoretical models. “You want to find out if your hypothesis can survive your 10 most dangerous experiments,” he says.

    Thursday, 25 May 2017

    China blocks online broadcast of computer go match

    BEIJING (AP) — Internet users outside China watched a computer defeat its national go champion, but few Chinese web surfers could see it.
    Censors blocked access to Tuesday's online broadcast by Google, which organized the game in a town west of Shanghai during a forum on artificial intelligence.
    The event got little coverage from Chinese newspapers and broadcasters, suggesting they may have received orders to avoid mentioning Google, which closed its China-based search engine in 2010 in a dispute over censorship and computer hacking.
    The official response to the game, a major event for go and artificial intelligence, reflects the conflict between the ruling Communist Party's technology ambitions and its insistence on controlling what its public can see, hear and read.
    The possible reason for suppressing coverage while allowing Google to organize the event in Wuzhen was unclear. Censorship orders to Chinese media are officially secret and government officials refuse to confirm whether online material is blocked.
    The event showcased AlphaGo, which beats top players at go, a 25 century-old game played with black and white stones on a chessboard-style grid. It is one of the last games that computers have yet to dominate.
    Tuesday's game took place in a hall where Chinese leaders hold the annual World Internet Conference, an event attended by global internet companies.
    Newspapers and websites reported AlphaGo's victory over Ke Jie in the first of three games they are to play this week.
    The reports were brief, even though Ke's post-game news conference was packed with scores of reporters, and none mentioned Google.
    There was no TV coverage, even though go has millions of fans in China and Ke, a 19-year-old prodigy, is a celebrity.
    Google says 60 million people in China watched online as AlphaGo played South Korea's go champion in March 2016.
    The Communist government encourages internet use for business and education but operates an elaborate system of monitoring and censorship.
    Censors block access to websites include social media and video-sharing websites such as Facebook and YouTube. Chinese internet companies are required to employ teams of censors to watch social media and remove banned material.
    China has the world's biggest population of internet users, with some 730 million people online by the end of last year, according to government data.
    Web surfers can get around online filters using virtual private networks, but Beijing has cracked down on use of those.
    Beijing's relationship with Google is especially prickly.
    The company, headquartered in Mountain View, California, opened a China-based search site in 2006 and cooperated with official censorship.
    That prompted complaints from human rights and other activists and free press groups such as Reporters Without Borders.
    In 2010, Google announced it no longer wanted to comply after hacking attacks on the company were traced to China.
    The company shut down its China search engine and visitors were automatically transferred to another Google service in Hong Kong.
    That stopped after Chinese authorities objected. Online filters slow access to the Hong Kong site enough to discourage many users.
    AlphaGo and Ke are to play their second game on Thursday.

    Computer wins 2nd game against Chinese go champion

    A computer beat China's top player of go, one of the last games machines have yet to master, for a second time Thursday in a competition authorities limited the Chinese public's ability to see.
    Ke Jie lost despite playing what Google's AlphaGo indicated was the best game any opponent has played against it, said Demis Hassabis, founder of the company that developed the program.
    AlphaGo defeated Ke, a 19-year-old prodigy, in their first game Tuesday during a forum organized by Google on artificial intelligence in Wuzhen, a town west of Shanghai. They play a final game Saturday.
    AlphaGo previously defeated European and South Korean champions, surprising players who had expected it to be at least a decade before computers could master the game.
    Internet users outside China could watch this week's games live but Chinese censors blocked most mainland web users from seeing the Google site carrying the feed. None of China's dozens of video sites carried the live broadcasts but a recording of Tuesday's game was available the following night on one popular site, Youku.com.
    State media reports on the games have been brief, possibly reflecting Beijing's antipathy toward Google, which closed its China-based search engine in 2010 following a dispute over censorship and computer hacking. Google says 60 million people in China watched online when AlphaGo played South Korea's go champion in March 2016.
    The official response to the match, a major event for the worlds of go and artificial intelligence, reflects the conflict between the ruling Communist Party's technology ambitions and its insistence on controlling what its public can see, hear and read.
    The government encourages internet use for business and education but tries to block access to material considered subversive.
    The possible reason for suppressing coverage while allowing Google to organize the event was unclear. Censorship orders to Chinese media are officially secret and government officials refuse to confirm whether online material is blocked.
    On Thursday, AlphaGo "thought that Ke Jie played perfectly" for the first 50 moves, Hassabis said at a news conference.
    "For the first roughly 100 moves, it is the closest game we have ever seen anyone play against the master version of AlphaGo," he said.
    Ke said the computer made unexpected moves after playing more methodically on Tuesday.
    "From the perspective of human beings, it stretched a little bit and I was surprised at some points," he said.
    "I also thought that I was very close to winning the match in the middle," Ke said. "I could feel my heart thumping. But maybe because I was too excited, I did some wrong or stupid moves. I guess that's the biggest weak point of human beings."
    Go players take turns putting white or black stones on a rectangular grid with 361 intersections, trying to capture territory and each other's pieces by surrounding them. The game is considered more difficult than chess for machines to master because the near-infinite number of possible positions requires intuition and flexibility.
    This week's games are taking place in a hall where Chinese leaders hold the annual World Internet Conference, an event attended by global internet companies.
    China has the world's biggest population of internet users, with some 730 million people online at the end of last year, according to government data.
    Censors block access to social media and video-sharing websites such as Facebook and YouTube. Internet companies are required to employ teams of censors to watch social media and remove banned material.
    Web surfers can get around online filters using virtual private networks, but Beijing has cracked down on use of those.

    How AR and computer vision will impact our lives for the better

    Technology moves at breakneck speed, and we now have more power in our pockets than we had in our homes in the 1990s. Augmented reality (AR) has been a fascinating concept of science fiction for decades, but many researchers think we’re finally getting close to making AR a reality thanks to advancements in computer vision.
    By definition, computer vision is a field that includes methods for acquiring, processing, analyzing, and understanding images and, in general, high-dimensional data from the real world in order to produce numerical or symbolic information, e.g., in the forms of decisions. In layman’s terms, computer visions allows machines to recognize and understand sight –  just as humans can.
    This means that with AR, you can process image and video sources to extract meaningful information and take action based on that.
    Human beings use sight to process, understand, and navigate the world around us. While much of this technology is still currently fairly rudimentary, we find ourselves with the ability to use AR and computer vision to one day significantly impact our everyday lives.
    According to Danny Lopez, the COO of Blippar – leading technology company specializing in augmented reality, artificial intelligence and computer vision – the challenge we have now is to find ways in the short, medium, and long term to apply and align AR to social good.
    Here are four ways Lopez feels AR might affect us in the future.
    1. Automated transportation
    We’re already seeing the beginnings of self-driving cars, though the vehicles are currently required to have a driver present at the wheel for safety. Despite these exciting developments, the technology isn’t perfect yet, and it will take a while for public acceptance to bring automated cars into widespread use. Another consideration won’t be that the car is driving itself, but its ability to be fully autonomous and protect all passengers. Currently, cars can only detect that there are people down the street. In the future, advanced cognition will help the car understand that those people are dangerous and you are in harm’s way.
    2. Healthcare
    Medical imaging has attracted increasing attention in recent years due to its vital component in healthcare applications. The advancement in computer vision – such as multimodal image fusion, medical image segmentation, image registration, computer-aided diagnosis, image annotation, and image-guided therapy – has opened up many new possibilities for revolutionizing healthcare.
    With literally millions of medical images indexed, AR and computer vision can match patterns of these images with similar ones from around the world to help doctors bring the best care to their patients.
    3. Education
    Initiatives in this field has been the improvement of the student’s experience through the use of computer vision. Integrating AR helps students with varying learning abilities.
    Additionally, computer vision applications may play a significant role in improving the effectiveness of traditional classroom tools – such as books and study materials – and aims to improve knowledge in specific areas.
    4. Training and manufacturing
    Product quality if a major concern for any manufacturing process. In every facility, the quality control division plays a big role. While these roles were traditionally done by humans, nowadays, it’s possible for computer vision to make quality control decisions.
    Cameras and lighting capture images which are then algorithmically compared to a predefined image or quality standard; thus eliminating human error.
    AR is also present in tasks that are too dangerous for humans alone including mining, fire-fighting, mine disposal and handling radioactive materials.
    The future
    It’s absolutely fascinating how far technology has advanced in a relatively short amount of time and according to Lopez, we’re “on the verge of it becoming mainstream.” But we’ve been on this tipping point for the last four years.
    As Lopez explains, “you can’t strongly scale AR if you don’t understand the reality in front of you. For this to happen, computer vision is absolutely necessary for AR immersive technology to really come to life.” But it’s coming.
    Over the last 25 years the progress of computing has learned to mimic more complex human behavior – and now it can mimic the whole set of human behaviors.

    Sunday, 21 May 2017

    The incredible life of DeepMind founder Demis Hassabis, the computer whiz who sold his AI lab to Google for £400 million



    Demis Hassabis, the 40-year-old cofounder of renowned artificial intelligence (AI) lab DeepMind, is recognised worldwide as one of the smartest thinkers in his field.
    Nicknamed the "superhero of artificial intelligence" by The Guardian, Hassabis is a former child chess prodigy with degrees in computer science and cognitive neuroscience from Cambridge and University College London respectively.
    Hassabis co-created the video game "Theme Park" game when he was just 17-years-old, before going on to found his own videogames company, and eventually DeepMind in 2010.
    In January 2014, Hassabis sold DeepMind to Google for £400 million in what is Google's largest European acquisition to date. The company made history last year when its self-learning AlphaGo agent beat a world champion at the notoriously difficult Chinese board game Go. Now DeepMind is turning its attention to applying its algorithms to areas that can benefit humanity, including healthcare and climate change.
    Hassabis was born in London July 27, 1976, to a Greek Cypriot father and a Chinese Singaporean mother.
    Hassabis is the eldest of three siblings and his parents are teachers. According to The Guardian, his sister is a pianist and composer, while his brother is a studying creative writing.
    "My parents are technophobes," he said during an interview with The Guardian. "They don't really like computers. They're kind of bohemian. My sister and brother both went the artistic route, too. None of them really went in for maths or science ... it's weird, I'm not quite sure where all this came from."
    Hassabis now has two young boys of his own.
    From an early age, he showed a natural aptitude for board games, particularly chess. Hassabis (bottom left) playing chess.DeepMind
    Hassabis first got interested in chess at the age of four while watching his father play a chess game against his uncle, according to Wired. Two weeks later, Hassabis was beating adults at the game.
    By the age of five, he was competing nationally and he won the London under-eights championships at the age of six. When he was nine he was captaining England's under-11 team.
    Hassabis bought his first computer when he was eight-years-old. It was a ZX Spectrum.
    Hassabis bought the machine with £200 that he won from a chess match.
    "The amazing thing about computers in those days is you could just start programming them," Hassabis told Wired. "I'd go with my dad to Foyles, and sit in the computer-programming department to learn how to give myself infinite lives in games. I intuitively understood that this was a magical device which you could unleash your creativity on."
    At the age of 13, Hassabis reached the rank of chess master. He was the second-highest-rated player in the world under 14 at the time. He finished his GCSEs when he was 14, two years ahead of everyone else in his class. He went on to take his maths A level when he was 15, followed by A levels in further maths, physics, and chemistry when he was 16. David Davies / PA Wire/Press Association Images He applied to the University of Cambridge and got a place but Cambridge wouldn't let him start because he would have only been 16 — so he took a gap year. He started his career in videogames at UK studio Bullfrog Productions when he was 15 after winning a competition for a job in Amiga Power magazine. At Bullfrog, he co-designed and led programming on "Theme Park," which challenges players to build a successful theme park.
    "The most fun I had in games was early in my career in the 90s," Hassabis told PCGamesN last July. "Especially at Bullfrog, I was lucky to be there at the most golden period it had. Maybe that has ever existed in the UK industry, if you look at the games it produced one after the other."
    During his time at the company he worked under legendary games designer Peter Molyneux, who was the founder of Bullfrog Productions.
    "I think we influenced each other a lot," Hassabis told PCGamesN. "We worked together very closely for a number of years — it's hard to say who [influenced who] more but it was a very important part of my life."
    "Theme Park" was released in 1994 when Hassabis was 17 and it went on to sell millions of copies. YouTube/Dubai Parks and Resorts Hassabis left Bullfrog Productions in 1994 to study computer science at Cambridge.
    Undergraduates at Cambridge were taught how to develop "narrow AI", which is able to learn how to perform specific tasks, but Hassabis was always more interested with developing "general AI", according to The Financial Times.
    He graduated from Queens' College Cambridge when he was 20 with a double first-class honours degree in 1997. After graduating in 1997, Hassabis worked at Lionhead Studios under Molyneux once again. YouTube/Singularity Videos
    At Lionhead, Hassabis worked on an early prototype version of the AI for iconic god game "Black & White".
    He left Lionhead around a year later to found his own videogames company.
    In 1998, Hassabis founded Elixir Studios, which produced award-winning games for global publishers such as Vivendi Universal and Microsoft.
    Elixir, which employed around 60 people at its peak, made AI simulation games such as "Republic: The Revolution" and the "Evil Genius, " which were both BAFTA-nominated.
    Acccording to The Financial Times, Hassabis sold a 5% stake in Elixir to Eidos, which created the Lara Croft "Tomb Raider" series. The stake was sold for £600,000, valuing the company at £12 million.
    After a decade in videogames startups, Hassabis returned to the world of academia at University College London (UCL) in 2005, where he completed a four-year PhD in cognitive neuroscience. UCL's Portico Building.Wikipedia / CC 3.0
    During his PhD, he sought to find inspiration in the human brain for new AI algorithms.
    His research on memory and imagination was listed in the top 10 scientific breakthroughs of the year by Science in 2007.
    In 2009, Hassabis was then awarded a Henry Wellcome postdoctoral research fellowship to continue his research at UCL for a further year.
    He also completed research stints in Boston, spending time at Harvard and... ... the Massachusetts Institute of Technology (MIT). In 2010, Hassabis founded what would go on to be his biggest company to date: DeepMind.
    DeepMind is a London-based startup that wants to "solve intelligence" and use it to "make the world a better place."
    The company is developing sophisticated self-learning algorithms that can excel at particular tasks when it is given a dataset to learn from. The algorithms are created by blending research and expertise from neuroscience and machine learning.
    So far, the algorithms have been used to defeat the best human player of Chinese board game Go and to help Google slash its enormous electricity bill. DeepMind is also applying its algorithms to a number of NHS projects.
    Hassabis cofounded DeepMind with childhood friend Mustafa Suleyman.
    Suleyman is head of applied AI at DeepMind and head of the DeepMind Health division, which is working with the NHS on a number of projects.
    Suleyman is also bright, having gained a place to study philosophy and theology at the University of Oxford. However, he dropped out in his second year when he was 19 and went on to launch the Muslim Youth Helpline. Suleyman went on to work as a policy officer for then Mayor of London, Ken Livingstone. Following that, he founded "Change Labs," which is a consultancy aimed at navigating complex problems.
    And New Zealander Shane Legg, who was another postdoc at UCL's Gatsby Computational Neuroscience Unit.
    Legg is chief scientist at DeepMind. He obtained his PhD from IDSIA in Switzerland, where he was supervised by Prof. Marcus Hutter, an expert on theoretical models of super intelligent machines.
    The 43-year-old works alongside Hassabis to lead DeepMind's research. Much of Legg's time is dedicated to hiring and deciding where DeepMind should focus its efforts next. Arguably more importantly, he also leads DeepMind's work on AI safety, which recently included developing a "big red button" to turn off machines when they start behaving in ways that humans don't want them to.
    Legg tends to stay out of the limelight and gives significantly fewer talks and far less quotes to journalists than his other cofounders. With the exception of this rare Bloomberg interview, you'll be hard pushed to find many stories about DeepMind that contain quotes from Legg.
    Hassabis's early investors include the likes of Tesla billionaire Elon Musk and Skype cofounder Jaan Tallinn.
    Musk explained his DeepMind investment to Vanity Fair earlier this year.
    "It gave me more visibility into the rate at which things were improving, and I think they're really improving at an accelerating rate, far faster than people realise," said Musk. "Mostly because in everyday life you don’t see robots walking around. Maybe your Roomba or something. But Roombas aren't going to take over the world."
    Not all of DeepMind's projects are well-known. Before DeepMind was acquired by Google, the company had an AI-powered fashion website called KITSEE. REUTERS / Shamil Zhumatov
    KITSEE used AI to recommend clothes to people that they could then go on and buy. The website also featured a range of articles about fashion that were produced by a team of DeepMind writers.
    KITSEE appears to have been abandoned around the time DeepMind was acquired by Google, suggesting the search giant may not have been interested in it.
    In December 2013, DeepMind revealed that it had made a breakthrough by training a piece of software to play Atari games at a superhuman level by only using the raw pixels on the screen as inputs. DeepMind was acquired by Google in 2014 for a reported £400 million when it had around 50 employees.
    Today, DeepMind sits under Google parent company Alphabet and employs around 400 staff in King's Cross, London. It also employs a small team at Google's headquarters in Mountain View, California, that are working on applying DeepMind's technology to Google products. However, it remains an independent organisation.
    As part of the Google acquisition, Hassabis and his cofounders made Google set up an AI ethics board. Who sits on that board has never been made public. In 2015, DeepMind made the front cover of Nature — something that many scientists dream of achieving — for its paper on how it created algorithms that could learn to master Atari arcade games. After cracking the Atari games, Hassabis and his rapidly-expanding team turned their attention to the ancient Chinese board game Go — a game that has been widely regarded as the holy grail in the AI community.
    Dating back more than 3,000 years, Go is a two-player board game that appears to be relatively simple on the surface — each player takes it in turns to lay a stone, with the objective being to surround the other player's pieces.
    However, the sheer number of potential moves on any given turn means that Go is in fact one of the most complex games humans have developed and AI scientists have been unable to master it for decades.
    AlphaGo learns by playing thousands of games against itself and gradually learning from its mistakes. 
    In March 2016, DeepMind pitched its Go-playing algorithm "AlphaGo" against world Go champion Lee SeDol. It was a five-game match that took place at the five star Four Seasons hotel in Seoul, South Korea.
    DeepMind won convincingly, taking four of the five games.
    Google cofounder and Alphabet president Sergey Brin came along to spectate.
    As did former Google CEO and Alphabet executive chairman Eric Schmidt.
    Lee Sedol was left stunned but he hasn't lost a game since. Ever since the defeat, he's been improving his game by practising on the AlphaGo algorithm. The main programmer on the AlphaGo algorithm was a man named David Silver, who was a fellow undergraduate with Hassabis at Cambridge.
    Silver and Hassabis also worked together at Elixir before DeepMind existed.
    "Dave and I have got a long history together," Hassabis told The Guardian in February 2016. "We used to dream about doing this in our lifetimes, so our 19-year-old selves would probably have been very relieved that we got here."
    DeepMind made the front cover of Nature for a second time in January 2016 for its work on the AlphaGo algorithm. AlphaGo is now on its way to China to take on some of the best players in the country where the game was invented. A number of AI experts, including Oxford professor Nick Bostrom, believe that DeepMind is currently winning the AI race, ahead of companies such as Facebook and Amazon. Nick Bostrom and Demis Hassabis.YouTube/Future of Life Hassabis has won a number of honours awards during his lifetime, including this one from The Asian Awards in May.
    In 2014, he won the Mullard Award from the Royal Society, and in 2016, he was ranked in Nature's "10 people that matter this year".
    In 2017, Time nominated him as one of the 100 most influential people in the world.

    Saturday, 20 May 2017

    How Google's strides in computer vision lead to Google Lens feature



    At the Google I/O conference on Wednesday, Google unveiled a new feature bringing computer vision capabilities to various products, starting with Google Assistant and Google Photos later this year.
    Called Google Lens, the feature helps you "understand what you're looking at" and take actions based on that information, CEO Sundar Pichai explained. For instance, you could point your phone at a flower and learn what kind of flower it is. You could point your phone at a restaurant and get contextual information such as its hours of operation. In another example, Pichai said, you could take a picture of a router and rather than typing in the password, "we can automatically do the hard work for you."
    As Google strives to integrate artificial intelligence into all of its products, Google Lens illustrates how far computer vision in particular has come. In fact, the vision error rate of computer vision algorithms is now better than the human error rate.
    Pichai called this "clearly at an inflection point with vision."
    "The fact computers can understand images and videos has profound implications for our core mission" of organizing the world's information, he added.


    Along with computer vision, Pichai gave an update on Google's speech-recognition capabilities, which now has a word error rate of 4.9 percent. He highlighted how advancements in speech recognition are influencing product design: Google initially planned to ship Google Home with eight microphones, but after applying deep learning, they were able to ship it with just two microphones and achieve the same quality. Thanks to deep learning, Google was able to recently announce support for multiple users in Google Home.
    With its focus now squarely on artificial intelligence, Google also announced Wednesday that it's launched Google.ai, a collection of all the efforts and teams from across the company related to AI. It focuses on three areas: State of the art research, tools and infrastructure, and applied AI.
    Google is "making impressive progress in applying machine learning and applying it across all products," Pichai said.
    In the meantime, the company continues to build its impressive reach across the globe. Several of Google's major products and platforms now have more than 1 billion users, Pichai said, including YouTube, Maps, Chrome, Gmail, Search, and Play. Meanwhile, Google Drive, launched five years ago, now has 800 million active users. Photos, launched two years ago, has more than 500 million active users.

    How hacked computer code allegedly helped a biker gang steal 150 Jeeps

    In a cross-border auto heist that resembles a scrapped plot from the “Fast and the Furious” franchise, nine members of a Tijuana-based bike...