Artificial intelligence software and a technique called deep learning could help Facebook understand its users and their data better.
Facebook is set to get an even better understanding of the 700 million people who use the social network to share details of their personal lives each day.
A new research group within the company is working on an emerging and powerful approach to artificial intelligence known as deep learning, which uses simulated networks of brain cells to process data. Applying this method to data shared on Facebook could allow for novel features and perhaps boost the company’s ad targeting.
Deep learning has shown potential as the basis for software that could work out the emotions or events described in text even if they aren’t explicitly referenced, recognize objects in photos, and make sophisticated predictions about people’s likely future behavior.
The eight-person group, known internally as the AI team, only recently started work, and details of its experiments are still secret. But Facebook’s chief technology officer, Mike Schroepfer, will say that one obvious way to use deep learning is to improve the news feed, the personalized list of recent updates he calls Facebook’s “killer app.”
Facebook already uses conventional machine learning techniques to prune the 1,500 updates that average Facebook users could possibly see down to 30 to 60 that are judged most likely to be important to them. Schroepfer says Facebook needs to get better at picking the best updates because its users are generating more data and using the social network in different ways.
Schroepfer told MIT Technology Review.
“The data set is increasing in size, people are getting more friends, and with the advent of mobile, people are online more frequently. It’s not that I look at my news feed once at the end of the day; I constantly pull out my phone while I’m waiting for my friend or I’m at the coffee shop. We have five minutes to really delight you.”
Shroepfer says deep learning could also be used to help people organize their photos or choose which is the best one to share on Facebook.
In looking into deep learning, Facebook follows its competitors Google and Microsoft, which have used the approach to impressive effect in the past year. Google has hired and acquired leading talent in the field (see “10 Breakthrough Technologies 2013: Deep Learning”), and last year it created software that taught itself to recognize cats and other objects by reviewing stills from YouTube videos. The underlying technology was later used to slash the error rate of Google’s voice recognition services (see “Google’s Virtual Brain Goes to Work”).
Meanwhile, researchers at Microsoft have used deep learning to build a system that translates speech from English to Mandarin Chinese in real time (see “Microsoft Brings Star Trek’s Voice Translator to Life”). Chinese Web giant Baidu also recently established a Silicon Valley research lab to work on deep learning.
Less complex forms of machine learning have underpinned some of the most useful features developed by major technology companies in recent years, such as spam detection systems and facial recognition in images. The largest companies have now begun investing heavily in deep learning because it can deliver significant gains over those more established techniques, says Elliot Turner, founder and CEO of AlchemyAPI, which rents access to its own deep learning software for text and images.
He says.
“Research into understanding images, text, and language has been going on for decades, but the typical improvement a new technique might offer was a fraction of a percent. In tasks like vision or speech, we’re seeing 30 percent-plus improvements with deep learning.”
The newer technique also allows much faster progress in training a new piece of software, says Turner.
Conventional forms of machine learning are slower because before data can be fed into learning software, experts must manually choose which features of it the software should pay attention to, and they must label the data to signify, for example, that certain images contain cars.
Deep learning systems can learn with much less human intervention because they can figure out for themselves which features of the raw data are most significant. They can even work on data that hasn’t been labeled, as Google’s cat-recognizing software did. Systems able to do that typically use software that simulates networks of brain cells, known as neural nets, to process data. They require more powerful collections of computers to run.
Facebook’s AI group will work on applications that can help the company’s products as well as on more general research that will be made public, says Srinivas Narayanan, an engineering manager at Facebook who’s helping to assemble the new group. He says one way Facebook can help advance deep learning is by drawing on its recent work creating new types of hardware and software to handle large data sets (see “Inside Facebook’s Not-So-Secret New Data Center”). He says.
“It’s both a software and a hardware problem together; the way you scale these networks requires very deep integration of the two.”
Facebook hired deep learning expert Marc’Aurelio Ranzato away from Google for its new group. Other members include Yaniv Taigman, cofounder of the facial recognition startup Face.com (see “When You’re Always a Familiar Face”); computer vision expert Lubomir Bourdev; and veteran Facebook engineer Keith Adams.
COMMENTARY: For several years, Google has been heavily involved in a branch of artificial intelligence or AI called deep learning. Deep-learning software attempts to mimic the activity in layers of neurons in the neocortex, the wrinkly 80 percent of the brain where thinking occurs. The software learns, in a very real sense, to recognize patterns in digital representations of sounds, images, and other data.
The basic idea—that software can simulate the neocortex’s large array of neurons in an artificial “neural network”—is decades old, and it has led to as many disappointments as breakthroughs. But because of improvements in mathematical formulas and increasingly powerful computers, computer scientists can now model many more layers of virtual neurons than ever before.
Jeff Dean, hired by Google CEO Larry Page to headup the development of large scale computers capable of handling the computing required for modern day artificial "neural networks" and deep learning, explains Google's research into deep learning and how this technology is appled across Google's product offerings:
With this greater depth, they are producing remarkable advances in speech and image recognition. Last June, a Google deep-learning system that had been shown 10 million images from YouTube videos proved almost twice as good as any previous image recognition effort at identifying objects such as cats. Google also used the technology to cut the error rate on speech recognition in its latest Android mobile software.
In October 2012, Microsoft chief research officer Rick Rashid (YouTube video below at 33:00) wowed attendees at a lecture in China with a demonstration of speech software that transcribed his spoken words into English text with an error rate of 7 percent, translated them into Chinese-language text, and then simulated his own voice uttering them in Mandarin. That same month, a team of three graduate students and two professors won a contest held by Merck to identify molecules that could lead to new drugs. The group used deep learning to zero in on the molecules most likely to bind to their targets.
Google in particular has become a magnet for deep learning and related AI talent. In March the company bought a startup cofounded by Geoffrey Hinton, a University of Toronto computer science professor who was part of the team that won the Merck contest. Hinton, who will split his time between the university and Google, says he plans to “take ideas out of this field and apply them to real problems” such as image recognition, search, and natural-language understanding, he says.
Extending deep learning into applications beyond speech and image recognition will require more conceptual and software breakthroughs, not to mention many more advances in processing power. And we probably won’t see machines we all agree can think for themselves for years, perhaps decades—if ever. But for now, says Peter Lee, head of Microsoft Research USA, “deep learning has reignited some of the grand challenges in artificial intelligence.”
Click Image To Enlarge
In June 2012, Google demonstrated one of the largest neural networks yet, with more than a billion connections. A team led by Stanford computer science professor Andrew Ng and Google Fellow Jeff Dean showed the system images from 10 million randomly selected YouTube videos. One simulated neuron in the software model fixated on images of cats. Others focused on human faces, yellow flowers, and other objects. And thanks to the power of deep learning, the system identified these discrete objects even though no humans had ever defined or labeled them.
What stunned some AI experts, though, was the magnitude of improvement in image recognition. The system correctly categorized objects and themes in the YouTube images 16 percent of the time. That might not sound impressive, but it was 70 percent better than previous methods. And, Dean notes, there were 22,000 categories to choose from; correctly slotting objects into some of them required, for example, distinguishing between two similar varieties of skate fish. That would have been challenging even for most humans. When the system was asked to sort the images into 1,000 more general categories, the accuracy rate jumped above 50 percent.
Big Data
Training the many layers of virtual neurons in the experiment took 16,000 computer processors—the kind of computing infrastructure that Google has developed for its search engine and other services. At least 80 percent of the recent advances in AI can be attributed to the availability of more computer power, reckons Dileep George, cofounder of the machine-learning startup Vicarious.
There’s more to it than the sheer size of Google’s data centers, though. Deep learning has also benefited from the company’s method of splitting computing tasks among many machines so they can be done much more quickly. That’s a technology Dean helped develop earlier in his 14-year career at Google. It vastly speeds up the training of deep-learning neural networks as well, enabling Google to run larger networks and feed a lot more data to them.
Already, deep learning has improved voice search on smartphones. Until last year, Google’s Android software used a method that misunderstood many words. But in preparation for a new release of Android last July, Dean and his team helped replace part of the speech system with one based on deep learning. Because the multiple layers of neurons allow for more precise training on the many variants of a sound, the system can recognize scraps of sound more reliably, especially in noisy environments such as subway platforms. Since it’s likelier to understand what was actually uttered, the result it returns is likelier to be accurate as well. Almost overnight, the number of errors fell by up to 25 percent—results so good that many reviewers now deem Android’s voice search smarter than Apple’s more famous Siri voice assistant.
Limitations of Deep Learning
For all the advances, not everyone thinks deep learning can move artificial intelligence toward something rivaling human intelligence. Some critics say deep learning and AI in general ignore too much of the brain’s biology in favor of brute-force computing.
One such critic is Jeff Hawkins, founder of Palm Computing, whose latest venture, Numenta, is developing a machine-learning system that is biologically inspired but does not use deep learning. Numenta’s system can help predict energy consumption patterns and the likelihood that a machine such as a windmill is about to fail. Hawkins, author of On Intelligence, a 2004 book on how the brain works and how it might provide a guide to building intelligent machines, says deep learning fails to account for the concept of time. Brains process streams of sensory data, he says, and human learning depends on our ability to recall sequences of patterns: when you watch a video of a cat doing something funny, it’s the motion that matters, not a series of still images like those Google used in its experiment. Hawkins says.
“Google’s attitude is: lots of data makes up for everything."
But if it doesn’t make up for everything, the computing resources a company like Google throws at these problems can’t be dismissed. They’re crucial, say deep-learning advocates, because the brain itself is still so much more complex than any of today’s neural networks. Hinton says.
“You need lots of computational resources to make the ideas work at all.”
What’s Next
Although Google is less than forthcoming about future applications, the prospects are intriguing. Clearly, better image search would help YouTube, for instance. And Dean says deep-learning models can use phoneme data from English to more quickly train systems to recognize the spoken sounds in other languages. It’s also likely that more sophisticated image recognition could make Google’s self-driving cars much better. Then there’s search and the ads that underwrite it. Both could see vast improvements from any technology that’s better and faster at recognizing what people are really looking for—maybe even before they realize it.
This is what intrigues Kurzweil, 65, who has long had a vision of intelligent machines. In high school, he wrote software that enabled a computer to create original music in various classical styles, which he demonstrated in a 1965 appearance on the TV show I’ve Got a Secret. Since then, his inventions have included several firsts—a print-to-speech reading machine, software that could scan and digitize printed text in any font, music synthesizers that could re-create the sound of orchestral instruments, and a speech recognition system with a large vocabulary.
Today, he envisions a “cybernetic friend” that listens in on your phone conversations, reads your e-mail, and tracks your every move—if you let it, of course—so it can tell you things you want to know even before you ask. This isn’t his immediate goal at Google, but it matches that of Google cofounder Sergey Brin, who said in the company’s early days that he wanted to build the equivalent of the sentient computer HAL in 2001: A Space Odyssey—except one that wouldn’t kill people.
For now, Kurzweil aims to help computers understand and even speak in natural language. “My mandate is to give computers enough understanding of natural language to do useful things—do a better job of search, do a better job of answering questions,” he says. Essentially, he hopes to create a more flexible version of IBM’s Watson, which he admires for its ability to understand Jeopardy!queries as quirky as “a long, tiresome speech delivered by a frothy pie topping.” (Watson’s correct answer: “What is a meringue harangue?”)
Kurzweil isn’t focused solely on deep learning, though he says his approach to speech recognition is based on similar theories about how the brain works. He wants to model the actual meaning of words, phrases, and sentences, including ambiguities that usually trip up computers. “I have an idea in mind of a graphical way to represent the semantic meaning of language,” he says.
That in turn will require a more comprehensive way to graph the syntax of sentences. Google is already using this kind of analysis to improve grammar in translations. Natural-language understanding will also require computers to grasp what we humans think of as common-sense meaning. For that, Kurzweil will tap into the Knowledge Graph, Google’s catalogue of some 700 million topics, locations, people, and more, plus billions of relationships among them. It was introduced last year as a way to provide searchers with answers to their queries, not just links.
Finally, Kurzweil plans to apply deep-learning algorithms to help computers deal with the “soft boundaries and ambiguities in language.” If all that sounds daunting, it is. “Natural-language understanding is not a goal that is finished at some point, any more than search,” he says. “That’s not a project I think I’ll ever finish.”
Though Kurzweil’s vision is still years from reality, deep learning is likely to spur other applications beyond speech and image recognition in the nearer term. For one, there’s drug discovery. The surprise victory by Hinton’s group in the Merck contest clearly showed the utility of deep learning in a field where few had expected it to make an impact.
That’s not all. Microsoft’s Peter Lee says there’s promising early research on potential uses of deep learning in machine vision—technologies that use imaging for applications such as industrial inspection and robot guidance. He also envisions personal sensors that deep neural networks could use to predict medical problems. And sensors throughout a city might feed deep-learning systems that could, for instance, predict where traffic jams might occur.
In a field that attempts something as profound as modeling the human brain, it’s inevitable that one technique won’t solve all the challenges. But for now, this one is leading the way in artificial intelligence.
Dean says.
“Deep learning is a really powerful metaphor for learning about the world.”
Conclusions
Although artificial intelligence has made some significant contributions in the science of deep learning, the technology is not even close to emulating the thinking processes that go in inside the human brain. Although IBM's Watson computer has demonstrated that it can compete and win against humans in a television game show like Jeopardy, we are a long way from developing systems that can match or outperform the human brain. Apple's personal assistant app SIRI uses voice commands to locate restaurants, movie theaters, Starbucks, and so forth, but results are often incorrect or way off course. In short, artificial intelligence is not really very intelligent. It is still computer programming that does a certain thing fairly well. When artificial intelligence has risen to the level of the HAL 9000 computer, and can think logically, intelligently, autonomosly, and learns and refines, then maybe we will take this science beyond the realm of science fiction to reality.
Courtesy of an article dated September 27, 2013 appearing in MIT Technology Review and an article dated April 23, 2013 appearing in MIT Technology Review
Comments