Tag Archives: Turing

#432487 Can We Make a Musical Turing Test?

As artificial intelligence advances, we’re encountering the same old questions. How much of what we consider to be fundamentally human can be reduced to an algorithm? Can we create something sufficiently advanced that people can no longer distinguish between the two? This, after all, is the idea behind the Turing Test, which has yet to be passed.

At first glance, you might think music is beyond the realm of algorithms. Birds can sing, and people can compose symphonies. Music is evocative; it makes us feel. Very often, our intense personal and emotional attachments to music are because it reminds us of our shared humanity. We are told that creative jobs are the least likely to be automated. Creativity seems fundamentally human.

But I think above all, we view it as reductionist sacrilege: to dissect beautiful things. “If you try to strangle a skylark / to cut it up, see how it works / you will stop its heart from beating / you will stop its mouth from singing.” A human musician wrote that; a machine might be able to string words together that are happy or sad; it might even be able to conjure up a decent metaphor from the depths of some neural network—but could it understand humanity enough to produce art that speaks to humans?

Then, of course, there’s the other side of the debate. Music, after all, has a deeply mathematical structure; you can train a machine to produce harmonics. “In the teachings of Pythagoras and his followers, music was inseparable from numbers, which were thought to be the key to the whole spiritual and physical universe,” according to Grout in A History of Western Music. You might argue that the process of musical composition cannot be reduced to a simple algorithm, yet musicians have often done so. Mozart, with his “Dice Music,” used the roll of a dice to decide how to order musical fragments; creativity through an 18th-century random number generator. Algorithmic music goes back a very long way, with the first papers on the subject from the 1960s.

Then there’s the techno-enthusiast side of the argument. iTunes has 26 million songs, easily more than a century of music. A human could never listen to and learn from them all, but a machine could. It could also memorize every note of Beethoven. Music can be converted into MIDI files, a nice chewable data format that allows even a character-by-character neural net you can run on your computer to generate music. (Seriously, even I could get this thing working.)

Indeed, generating music in the style of Bach has long been a test for AI, and you can see neural networks gradually learn to imitate classical composers while trying to avoid overfitting. When an algorithm overfits, it essentially starts copying the existing music, rather than being inspired by it but creating something similar: a tightrope the best human artists learn to walk. Creativity doesn’t spring from nowhere; even maverick musical geniuses have their influences.

Does a machine have to be truly ‘creative’ to produce something that someone would find valuable? To what extent would listeners’ attitudes change if they thought they were hearing a human vs. an AI composition? This all suggests a musical Turing Test. Of course, it already exists. In fact, it’s run out of Dartmouth, the school that hosted that first, seminal AI summer conference. This year, the contest is bigger than ever: alongside the PoetiX, LimeriX and LyriX competitions for poetry and lyrics, there’s a DigiKidLit competition for children’s literature (although you may have reservations about exposing your children to neural-net generated content… it can get a bit surreal).

There’s also a pair of musical competitions, including one for original compositions in different genres. Key genres and styles are represented by Charlie Parker for Jazz and the Bach chorales for classical music. There’s also a free composition, and a contest where a human and an AI try to improvise together—the AI must respond to a human spontaneously, in real time, and in a musically pleasing way. Quite a challenge! In all cases, if any of the generated work is indistinguishable from human performers, the neural net has passed the Turing Test.

Did they? Here’s part of 2017’s winning sonnet from Charese Smiley and Hiroko Bretz:

The large cabin was in total darkness.
Come marching up the eastern hill afar.
When is the clock on the stairs dangerous?
Everything seemed so near and yet so far.
Behind the wall silence alone replied.
Was, then, even the staircase occupied?
Generating the rhymes is easy enough, the sentence structure a little trickier, but what’s impressive about this sonnet is that it sticks to a single topic and appears to be a more coherent whole. I’d guess they used associated “lexical fields” of similar words to help generate something coherent. In a similar way, most of the more famous examples of AI-generated music still involve some amount of human control, even if it’s editorial; a human will build a song around an AI-generated riff, or select the most convincing Bach chorale from amidst many different samples.

We are seeing strides forward in the ability of AI to generate human voices and human likenesses. As the latter example shows, in the fake news era people have focused on the dangers of this tech– but might it also be possible to create a virtual performer, trained on a dataset of their original music? Did you ever want to hear another Beatles album, or jam with Miles Davis? Of course, these things are impossible—but could we create a similar experience that people would genuinely value? Even, to the untrained eye, something indistinguishable from the real thing?

And if it did measure up to the real thing, what would this mean? Jaron Lanier is a fascinating technology writer, a critic of strong AI, and a believer in the power of virtual reality to change the world and provide truly meaningful experiences. He’s also a composer and a musical aficionado. He pointed out in a recent interview that translation algorithms, by reducing the amount of work translators are commissioned to do, have, in some sense, profited from stolen expertise. They were trained on huge datasets purloined from human linguists and translators. If you can train an AI on someone’s creative output and it produces new music, who “owns” it?

Although companies that offer AI music tools are starting to proliferate, and some groups will argue that the musical Turing test has been passed already, AI-generated music is hardly racing to the top of the pop charts just yet. Even as the line between human-composed and AI-generated music starts to blur, there’s still a gulf between the average human and musical genius. In the next few years, we’ll see how far the current techniques can take us. It may be the case that there’s something in the skylark’s song that can’t be generated by machines. But maybe not, and then this song might need an extra verse.

Image Credit: d1sk / Shutterstock.com Continue reading

Posted in Human Robots

#432467 Dungeons and Dragons, Not Chess and Go: ...

Everyone had died—not that you’d know it, from how they were laughing about their poor choices and bad rolls of the dice. As a social anthropologist, I study how people understand artificial intelligence (AI) and our efforts towards attaining it; I’m also a life-long fan of Dungeons and Dragons (D&D), the inventive fantasy roleplaying game. During a recent quest, when I was playing an elf ranger, the trainee paladin (or holy knight) acted according to his noble character, and announced our presence at the mouth of a dragon’s lair. The results were disastrous. But while success in D&D means “beating the bad guy,” the game is also a creative sandbox, where failure can count as collective triumph so long as you tell a great tale.

What does this have to do with AI? In computer science, games are frequently used as a benchmark for an algorithm’s “intelligence.” The late Robert Wilensky, a professor at the University of California, Berkeley and a leading figure in AI, offered one reason why this might be. Computer scientists “looked around at who the smartest people were, and they were themselves, of course,” he told the authors of Compulsive Technology: Computers as Culture (1985). “They were all essentially mathematicians by training, and mathematicians do two things—they prove theorems and play chess. And they said, hey, if it proves a theorem or plays chess, it must be smart.” No surprise that demonstrations of AI’s “smarts” have focused on the artificial player’s prowess.

Yet the games that get chosen—like Go, the main battlefield for Google DeepMind’s algorithms in recent years—tend to be tightly bounded, with set objectives and clear paths to victory or defeat. These experiences have none of the open-ended collaboration of D&D. Which got me thinking: do we need a new test for intelligence, where the goal is not simply about success, but storytelling? What would it mean for an AI to “pass” as human in a game of D&D? Instead of the Turing test, perhaps we need an elf ranger test?

Of course, this is just a playful thought experiment, but it does highlight the flaws in certain models of intelligence. First, it reveals how intelligence has to work across a variety of environments. D&D participants can inhabit many characters in many games, and the individual player can “switch” between roles (the fighter, the thief, the healer). Meanwhile, AI researchers know that it’s super difficult to get a well-trained algorithm to apply its insights in even slightly different domains—something that we humans manage surprisingly well.

Second, D&D reminds us that intelligence is embodied. In computer games, the bodily aspect of the experience might range from pressing buttons on a controller in order to move an icon or avatar (a ping-pong paddle; a spaceship; an anthropomorphic, eternally hungry, yellow sphere), to more recent and immersive experiences involving virtual-reality goggles and haptic gloves. Even without these add-ons, games can still produce biological responses associated with stress and fear (if you’ve ever played Alien: Isolation you’ll understand). In the original D&D, the players encounter the game while sitting around a table together, feeling the story and its impact. Recent research in cognitive science suggests that bodily interactions are crucial to how we grasp more abstract mental concepts. But we give minimal attention to the embodiment of artificial agents, and how that might affect the way they learn and process information.

Finally, intelligence is social. AI algorithms typically learn through multiple rounds of competition, in which successful strategies get reinforced with rewards. True, it appears that humans also evolved to learn through repetition, reward and reinforcement. But there’s an important collaborative dimension to human intelligence. In the 1930s, the psychologist Lev Vygotsky identified the interaction of an expert and a novice as an example of what became called “scaffolded” learning, where the teacher demonstrates and then supports the learner in acquiring a new skill. In unbounded games, this cooperation is channelled through narrative. Games of It among small children can evolve from win/lose into attacks by terrible monsters, before shifting again to more complex narratives that explain why the monsters are attacking, who is the hero, and what they can do and why—narratives that aren’t always logical or even internally compatible. An AI that could engage in social storytelling is doubtless on a surer, more multifunctional footing than one that plays chess; and there’s no guarantee that chess is even a step on the road to attaining intelligence of this sort.

In some ways, this failure to look at roleplaying as a technical hurdle for intelligence is strange. D&D was a key cultural touchstone for technologists in the 1980s and the inspiration for many early text-based computer games, as Katie Hafner and Matthew Lyon point out in Where Wizards Stay up Late: The Origins of the Internet (1996). Even today, AI researchers who play games in their free time often mention D&D specifically. So instead of beating adversaries in games, we might learn more about intelligence if we tried to teach artificial agents to play together as we do: as paladins and elf rangers.

This article was originally published at Aeon and has been republished under Creative Commons.

Image Credit:Benny Mazur/Flickr / CC BY 2.0 Continue reading

Posted in Human Robots

#431928 How Fast Is AI Progressing? Stanford’s ...

When? This is probably the question that futurists, AI experts, and even people with a keen interest in technology dread the most. It has proved famously difficult to predict when new developments in AI will take place. The scientists at the Dartmouth Summer Research Project on Artificial Intelligence in 1956 thought that perhaps two months would be enough to make “significant advances” in a whole range of complex problems, including computers that can understand language, improve themselves, and even understand abstract concepts.
Sixty years later, and these problems are not yet solved. The AI Index, from Stanford, is an attempt to measure how much progress has been made in artificial intelligence.
The index adopts a unique approach, and tries to aggregate data across many regimes. It contains Volume of Activity metrics, which measure things like venture capital investment, attendance at academic conferences, published papers, and so on. The results are what you might expect: tenfold increases in academic activity since 1996, an explosive growth in startups focused around AI, and corresponding venture capital investment. The issue with this metric is that it measures AI hype as much as AI progress. The two might be correlated, but then again, they may not.
The index also scrapes data from the popular coding website Github, which hosts more source code than anyone in the world. They can track the amount of AI-related software people are creating, as well as the interest levels in popular machine learning packages like Tensorflow and Keras. The index also keeps track of the sentiment of news articles that mention AI: surprisingly, given concerns about the apocalypse and an employment crisis, those considered “positive” outweigh the “negative” by three to one.
But again, this could all just be a measure of AI enthusiasm in general.
No one would dispute the fact that we’re in an age of considerable AI hype, but the progress of AI is littered by booms and busts in hype, growth spurts that alternate with AI winters. So the AI Index attempts to track the progress of algorithms against a series of tasks. How well does computer vision perform at the Large Scale Visual Recognition challenge? (Superhuman at annotating images since 2015, but they still can’t answer questions about images very well, combining natural language processing and image recognition). Speech recognition on phone calls is almost at parity.
In other narrow fields, AIs are still catching up to humans. Translation might be good enough that you can usually get the gist of what’s being said, but still scores poorly on the BLEU metric for translation accuracy. The AI index even keeps track of how well the programs can do on the SAT test, so if you took it, you can compare your score to an AI’s.
Measuring the performance of state-of-the-art AI systems on narrow tasks is useful and fairly easy to do. You can define a metric that’s simple to calculate, or devise a competition with a scoring system, and compare new software with old in a standardized way. Academics can always debate about the best method of assessing translation or natural language understanding. The Loebner prize, a simplified question-and-answer Turing Test, recently adopted Winograd Schema type questions, which rely on contextual understanding. AI has more difficulty with these.
Where the assessment really becomes difficult, though, is in trying to map these narrow-task performances onto general intelligence. This is hard because of a lack of understanding of our own intelligence. Computers are superhuman at chess, and now even a more complex game like Go. The braver predictors who came up with timelines thought AlphaGo’s success was faster than expected, but does this necessarily mean we’re closer to general intelligence than they thought?
Here is where it’s harder to track progress.
We can note the specialized performance of algorithms on tasks previously reserved for humans—for example, the index cites a Nature paper that shows AI can now predict skin cancer with more accuracy than dermatologists. We could even try to track one specific approach to general AI; for example, how many regions of the brain have been successfully simulated by a computer? Alternatively, we could simply keep track of the number of professions and professional tasks that can now be performed to an acceptable standard by AI.

“We are running a race, but we don’t know how to get to the endpoint, or how far we have to go.”

Progress in AI over the next few years is far more likely to resemble a gradual rising tide—as more and more tasks can be turned into algorithms and accomplished by software—rather than the tsunami of a sudden intelligence explosion or general intelligence breakthrough. Perhaps measuring the ability of an AI system to learn and adapt to the work routines of humans in office-based tasks could be possible.
The AI index doesn’t attempt to offer a timeline for general intelligence, as this is still too nebulous and confused a concept.
Michael Woodridge, head of Computer Science at the University of Oxford, notes, “The main reason general AI is not captured in the report is that neither I nor anyone else would know how to measure progress.” He is concerned about another AI winter, and overhyped “charlatans and snake-oil salesmen” exaggerating the progress that has been made.
A key concern that all the experts bring up is the ethics of artificial intelligence.
Of course, you don’t need general intelligence to have an impact on society; algorithms are already transforming our lives and the world around us. After all, why are Amazon, Google, and Facebook worth any money? The experts agree on the need for an index to measure the benefits of AI, the interactions between humans and AIs, and our ability to program values, ethics, and oversight into these systems.
Barbra Grosz of Harvard champions this view, saying, “It is important to take on the challenge of identifying success measures for AI systems by their impact on people’s lives.”
For those concerned about the AI employment apocalypse, tracking the use of AI in the fields considered most vulnerable (say, self-driving cars replacing taxi drivers) would be a good idea. Society’s flexibility for adapting to AI trends should be measured, too; are we providing people with enough educational opportunities to retrain? How about teaching them to work alongside the algorithms, treating them as tools rather than replacements? The experts also note that the data suffers from being US-centric.
We are running a race, but we don’t know how to get to the endpoint, or how far we have to go. We are judging by the scenery, and how far we’ve run already. For this reason, measuring progress is a daunting task that starts with defining progress. But the AI index, as an annual collection of relevant information, is a good start.
Image Credit: Photobank gallery / Shutterstock.com Continue reading

Posted in Human Robots

#431869 When Will We Finally Achieve True ...

The field of artificial intelligence goes back a long way, but many consider it was officially born when a group of scientists at Dartmouth College got together for a summer, back in 1956. Computers had, over the last few decades, come on in incredible leaps and bounds; they could now perform calculations far faster than humans. Optimism, given the incredible progress that had been made, was rational. Genius computer scientist Alan Turing had already mooted the idea of thinking machines just a few years before. The scientists had a fairly simple idea: intelligence is, after all, just a mathematical process. The human brain was a type of machine. Pick apart that process, and you can make a machine simulate it.
The problem didn’t seem too hard: the Dartmouth scientists wrote, “We think that a significant advance can be made in one or more of these problems if a carefully selected group of scientists work on it together for a summer.” This research proposal, by the way, contains one of the earliest uses of the term artificial intelligence. They had a number of ideas—maybe simulating the human brain’s pattern of neurons could work and teaching machines the abstract rules of human language would be important.
The scientists were optimistic, and their efforts were rewarded. Before too long, they had computer programs that seemed to understand human language and could solve algebra problems. People were confidently predicting there would be a human-level intelligent machine built within, oh, let’s say, the next twenty years.
It’s fitting that the industry of predicting when we’d have human-level intelligent AI was born at around the same time as the AI industry itself. In fact, it goes all the way back to Turing’s first paper on “thinking machines,” where he predicted that the Turing Test—machines that could convince humans they were human—would be passed in 50 years, by 2000. Nowadays, of course, people are still predicting it will happen within the next 20 years, perhaps most famously Ray Kurzweil. There are so many different surveys of experts and analyses that you almost wonder if AI researchers aren’t tempted to come up with an auto reply: “I’ve already predicted what your question will be, and no, I can’t really predict that.”
The issue with trying to predict the exact date of human-level AI is that we don’t know how far is left to go. This is unlike Moore’s Law. Moore’s Law, the doubling of processing power roughly every couple of years, makes a very concrete prediction about a very specific phenomenon. We understand roughly how to get there—improved engineering of silicon wafers—and we know we’re not at the fundamental limits of our current approach (at least, not until you’re trying to work on chips at the atomic scale). You cannot say the same about artificial intelligence.
Common Mistakes
Stuart Armstrong’s survey looked for trends in these predictions. Specifically, there were two major cognitive biases he was looking for. The first was the idea that AI experts predict true AI will arrive (and make them immortal) conveniently just before they’d be due to die. This is the “Rapture of the Nerds” criticism people have leveled at Kurzweil—his predictions are motivated by fear of death, desire for immortality, and are fundamentally irrational. The ability to create a superintelligence is taken as an article of faith. There are also criticisms by people working in the AI field who know first-hand the frustrations and limitations of today’s AI.
The second was the idea that people always pick a time span of 15 to 20 years. That’s enough to convince people they’re working on something that could prove revolutionary very soon (people are less impressed by efforts that will lead to tangible results centuries down the line), but not enough for you to be embarrassingly proved wrong. Of the two, Armstrong found more evidence for the second one—people were perfectly happy to predict AI after they died, although most didn’t, but there was a clear bias towards “15–20 years from now” in predictions throughout history.
Measuring Progress
Armstrong points out that, if you want to assess the validity of a specific prediction, there are plenty of parameters you can look at. For example, the idea that human-level intelligence will be developed by simulating the human brain does at least give you a clear pathway that allows you to assess progress. Every time we get a more detailed map of the brain, or successfully simulate another part of it, we can tell that we are progressing towards this eventual goal, which will presumably end in human-level AI. We may not be 20 years away on that path, but at least you can scientifically evaluate the progress.
Compare this to those that say AI, or else consciousness, will “emerge” if a network is sufficiently complex, given enough processing power. This might be how we imagine human intelligence and consciousness emerged during evolution—although evolution had billions of years, not just decades. The issue with this is that we have no empirical evidence: we have never seen consciousness manifest itself out of a complex network. Not only do we not know if this is possible, we cannot know how far away we are from reaching this, as we can’t even measure progress along the way.
There is an immense difficulty in understanding which tasks are hard, which has continued from the birth of AI to the present day. Just look at that original research proposal, where understanding human language, randomness and creativity, and self-improvement are all mentioned in the same breath. We have great natural language processing, but do our computers understand what they’re processing? We have AI that can randomly vary to be “creative,” but is it creative? Exponential self-improvement of the kind the singularity often relies on seems far away.
We also struggle to understand what’s meant by intelligence. For example, AI experts consistently underestimated the ability of AI to play Go. Many thought, in 2015, it would take until 2027. In the end, it took two years, not twelve. But does that mean AI is any closer to being able to write the Great American Novel, say? Does it mean it’s any closer to conceptually understanding the world around it? Does it mean that it’s any closer to human-level intelligence? That’s not necessarily clear.
Not Human, But Smarter Than Humans
But perhaps we’ve been looking at the wrong problem. For example, the Turing test has not yet been passed in the sense that AI cannot convince people it’s human in conversation; but of course the calculating ability, and perhaps soon the ability to perform other tasks like pattern recognition and driving cars, far exceed human levels. As “weak” AI algorithms make more decisions, and Internet of Things evangelists and tech optimists seek to find more ways to feed more data into more algorithms, the impact on society from this “artificial intelligence” can only grow.
It may be that we don’t yet have the mechanism for human-level intelligence, but it’s also true that we don’t know how far we can go with the current generation of algorithms. Those scary surveys that state automation will disrupt society and change it in fundamental ways don’t rely on nearly as many assumptions about some nebulous superintelligence.
Then there are those that point out we should be worried about AI for other reasons. Just because we can’t say for sure if human-level AI will arrive this century, or never, it doesn’t mean we shouldn’t prepare for the possibility that the optimistic predictors could be correct. We need to ensure that human values are programmed into these algorithms, so that they understand the value of human life and can act in “moral, responsible” ways.
Phil Torres, at the Project for Future Human Flourishing, expressed it well in an interview with me. He points out that if we suddenly decided, as a society, that we had to solve the problem of morality—determine what was right and wrong and feed it into a machine—in the next twenty years…would we even be able to do it?
So, we should take predictions with a grain of salt. Remember, it turned out the problems the AI pioneers foresaw were far more complicated than they anticipated. The same could be true today. At the same time, we cannot be unprepared. We should understand the risks and take our precautions. When those scientists met in Dartmouth in 1956, they had no idea of the vast, foggy terrain before them. Sixty years later, we still don’t know how much further there is to go, or how far we can go. But we’re going somewhere.
Image Credit: Ico Maker / Shutterstock.com Continue reading

Posted in Human Robots

#431392 What AI Can Now Do Is Remarkable—But ...

Major websites all over the world use a system called CAPTCHA to verify that someone is indeed a human and not a bot when entering data or signing into an account. CAPTCHA stands for the “Completely Automated Public Turing test to tell Computers and Humans Apart.” The squiggly letters and numbers, often posted against photographs or textured backgrounds, have been a good way to foil hackers. They are annoying but effective.
The days of CAPTCHA as a viable line of defense may, however, be numbered.
Researchers at Vicarious, a Californian artificial intelligence firm funded by Amazon founder Jeff Bezos and Facebook’s Mark Zuckerberg, have just published a paper documenting how they were able to defeat CAPTCHA using new artificial intelligence techniques. Whereas today’s most advanced artificial intelligence (AI) technologies use neural networks that require massive amounts of data to learn from, sometimes millions of examples, the researchers said their system needed just five training steps to crack Google’s reCAPTCHA technology. With this, they achieved a 67 percent success rate per character—reasonably close to the human accuracy rate of 87 percent. In answering PayPal and Yahoo CAPTCHAs, the system achieved an accuracy rate of greater than 50 percent.
The CAPTCHA breakthrough came hard on the heels of another major milestone from Google’s DeepMind team, the people who built the world’s best Go-playing system. DeepMind built a new artificial-intelligence system called AlphaGo Zero that taught itself to play the game at a world-beating level with minimal training data, mainly using trial and error—in a fashion similar to how humans learn.
Both playing Go and deciphering CAPTCHAs are clear examples of what we call narrow AI, which is different from artificial general intelligence (AGI)—the stuff of science fiction. Remember R2-D2 of Star Wars, Ava from Ex Machina, and Samantha from Her? They could do many things and learned everything they needed on their own.
Narrow AI technologies are systems that can only perform one specific type of task. For example, if you asked AlphaGo Zero to learn to play Monopoly, it could not, even though that is a far less sophisticated game than Go. If you asked the CAPTCHA cracker to learn to understand a spoken phrase, it would not even know where to start.
To date, though, even narrow AI has been difficult to build and perfect. To perform very elementary tasks such as determining whether an image is of a cat or a dog, the system requires the development of a model that details exactly what is being analyzed and massive amounts of data with labeled examples of both. The examples are used to train the AI systems, which are modeled on the neural networks in the brain, in which the connections between layers of neurons are adjusted based on what is observed. To put it simply, you tell an AI system exactly what to learn, and the more data you give it, the more accurate it becomes.
The methods that Vicarious and Google used were different; they allowed the systems to learn on their own, albeit in a narrow field. By making their own assumptions about what the training model should be and trying different permutations until they got the right results, they were able to teach themselves how to read the letters in a CAPTCHA or to play a game.
This blurs the line between narrow AI and AGI and has broader implications in robotics and virtually any other field in which machine learning in complex environments may be relevant.
Beyond visual recognition, the Vicarious breakthrough and AlphaGo Zero success are encouraging scientists to think about how AIs can learn to do things from scratch. And this brings us one step closer to coexisting with classes of AIs and robots that can learn to perform new tasks that are slight variants on their previous tasks—and ultimately the AGI of science fiction.
So R2-D2 may be here sooner than we expected.
This article was originally published by The Washington Post. Read the original article here.
Image Credit: Zapp2Photo / Shutterstock.com Continue reading

Posted in Human Robots