Tag Archives: deep learning

AI turns the corner?

Posted on by
15

Is the latest version of AlphaZero showing signs of human style intuition and creativity?

The current AlphaZero is a more generalised version of the program, produced by Demis Hassabis and his team, that beat top human players of Go for the first time. The new version, presented briefly in Science magazine, is able to learn a range of different games; besides Go it learned chess and shogi, and apparently reached world-class play in all of them.

The Science article, by David Silver et al, modestly says this achievement is an important step towards a fully-general game-playing program, but press reports went further, claiming that in chess particularly AlphaZero showed more human traits than any previous system. It reinvented human strategies and sacrificed pieces for advantage fairly readily, the way human players do; chess commentators said that its play seemed to have new qualities of insight and intuition.

This is somewhat unexpected, because so far as I can tell the latest version is in fact not a radical change from its predecessors; in essence it uses the same clever combination of appropriate algorithms with a deep neural network, simply applying them more generally. It does appear that the approach has proved more widely powerful than we might have expected, but it is no more human in nature than the earlier versions and does not embody any new features copied from real brains. It learns its games from scratch, with only the rules to go on, playing out games against itself repeatedly in order to find what works. This is not much like the way humans learn chess; I think you would probably die of boredom after a few hundred games, and even if you survived, without some instruction and guidance you would probably never learn to be a good player, let alone a superlative one. However, running through possible games in one’s mind may be quite like what a good human player does when trying to devise new strategies.

The key point for me is that although the new program is far more general in application, it still only operates in the well-defined and simple worlds provided by rule-governed games. To be anything like human, it needs to display the ability to deal with the heterogenous and undefinable world of real life. That is still far distant (Hassabis himself has displayed an awareness of the scale of the problem, warning against releasing self-driving cars on to real roads prematurely), though I don’t altogether rule out the possibility that we are now moving perceptibly in the right direction.

Someone who might deny that is Gary Marcus, who in a recent Nautilus piece set out his view that deep learning is simply not enough. It needs, he says, to be supplemented by other tools, and in particular it needs symbol manipulation.

To me this is confusing, because I naturally interpret ‘symbol manipulation’ as being pretty much a synonym for Turing style computation. That’s the bedrock of any conventional computer, so it seems odd to say we need to add it. I suppose Marcus is using the word ‘symbol’ in a different sense. The ones and zeroes shuffled around by a Turing machine are meaningless to the machine. We assign a meaning to the input and so devise the manipulations that the output can be given an interpretation which makes it useful or interesting to us, but the machine itself knows nothing of that. Marcus perhaps means that we need a new generation of machines that can handle symbols according to their meanings.

If that’s it, then few would disagree that that is one of the ultimate aims. Those who espouse deep learning techniques merely think that those methods may in due course lead to a machine that handles meanings in the true sense; at some stage the system will come up with the unknown general strategy that enables it to get meaningful and use symbols the way a human would. Marcus presumably thinks that is hopeless optimism, on the scale of those people who think any system that is sufficiently complex will naturally attain self-awareness.

Since we don’t have much of an idea how the miracle of meaning might happen it is indeed optimistic to think we’re on the road towards it. How can a machine bootstrap itself into true ‘symbol manipulation’ without some kind of help? But we know that the human brain must have done just that at some stage in evolution – and indeed each of our brains seem to have done it again during our development. It has got to be possible. Optimistic yes, hopeless – maybe not.

A Case for Human Thinking

Posted on by
8

neural-netInteresting piece here reviewing the way some modern machine learning systems are unfathomable. This is because they learn how to do what they do, rather than being set up with a program, so there is no reassuring algorithm – no set of instructions that tells us how they work. In fact they way they make their decisions may be impossible to grasp properly even if we know all the details because it just exceeds in brute complexity what we can ever get our minds around.

This is not really new. Neural nets that learn for themselves have always been a bit inscrutable. One problem with this is brittleness: when the system fails it may not fail in ways that are safe and manageable, but disastrously. This old problem is becoming more important mainly because new approaches to deep machine learning are doing so well; all of a sudden we seem to be getting a rush of new systems that really work effectively at quite complex real world tasks. The problems are no longer academic.

Brittle behaviour may come about when the system learns its task from a limited data set. It does not understand the data and is simply good at picking out correlations, so sometimes it may pick out features of the original data set that work well within that set, and perhaps even work well on quite a lot of new real world data, but don’t really capture what’s important. The program is meant to check whether a station platform is dangerously full of people, for example; in the set of pictures provided it finds that all it needs to do is examine the white platform area and check how dark it is. The more people there are, the darker it looks. This turns out to work quite well in real life, too. Then summer comes and people start wearing light coloured clothes…

There are ways to cope with this. We could build in various safeguards. We could make sure we use big and realistic datasets for training or perhaps allow learning to continue in real world contexts. Or we could just decide never to use a system that doesn’t have an algorithm we can examine; but there would be a price to pay in terms of efficiency for that; it might even be that we would have to give up on certain things that can only be effectively automated with relatively sophisticated deep learning methods. We’re told that the EU contemplates a law embodying a right to explanations of how software works. To philosophers I think this must sound like a marvellous new gravy train, as there will obviously be a need to adjudicate what counts as an adequate explanation, a notoriously problematic issue. (I am available as a witness in any litigation for reasonable hourly fees.)

The article points out that the incomprehensibility of neural network-based systems is in some ways really quite like the incomprehensibility of the good old human brain. Why wouldn’t it be? After all, neural nets were based on the brain. Now it’s true that even in the beginning they were very rough approximations of real neurology and in practical modern systems the neural qualities of neural nets are little more than a polite fiction. Still, perhaps there are properties shared by all learning systems?

One reason deep learning may run into problems is the difficulty AI always has in dealing with relevance.  The ability to spot relevance no doubt helps the human brain check whether it is learning about the right kind of thing, but it has always been difficult to work out quite how our brains do it, and this might mean an essential element is missing from AI approaches.

It is tempting, though, to think that this is in part another manifestation of the fact that AI systems get trained on limited data sets. Maybe the radical answer is to stop feeding them tailored data sets and let  our robots live in the real world; in other words, if we want reliable deep learning perhaps our robots have to roam free and replicate the wider human experience of the world at large? To date the project of creating human-style cognition has been in some sense motivated by mere curiosity (and yes, by the feeling that it would be pretty cool to have a robot pal) ; are we seeing here the outline of an argument that human-style AGI might actually be the answer to genuine engineering problems?

What about those explanations? Instead of retaining philosophers and lawyers to argue the case, could we think about building in a new module to our systems, one that keeps overall track of the AI and can report the broad currents of activity within it? It wouldn’t be perfect but it might give us broad clues as to why the system was making the decisions it was, and even allow us to delicately feed in some guidance. Doesn’t such a module start to sound like, well, consciousness? Could it be that we are beginning to see the outline of the rationales behind some of God’s design choices?

Preparing the triumph of brute force?

Posted on by
12

ray kurzweilThe Guardian had a piece recently which was partly a profile of Ray Kurzweil, and partly about the way Google seems to have gone on a buying spree, snapping up experts on machine learning and robotics – with Kurzweil himself made Director of Engineering.

The problem with Ray Kurzweil is that he is two people. There is Ray Kurzweil the competent and genuinely gifted innovator, a man we hear little from: and then there’s Ray Kurzweil the motor-mouth, prophet of the Singularity, aspirant immortal, and gushing fountain of optimistic predictions. The Guardian piece praises his record of prediction, rather oddly quoting in support his prediction that by the year 2000 paraplegics would be walking with robotic leg prostheses – something that in 2014 has still not happened. That perhaps does provide a clue to the Kurzweil method: if you issue thousands of moderately plausible predictions, some will pay off. A doubtless-apocryphal story has it that at AI conferences people play the Game of Kurzweil. Players take turns to offer a Kurzweilian prediction (by 2020 there will be a restaurant where sensors sniff your breath and the ideal meal is got ready without you needing to order; by 2050 doctors will routinely use special machines to selectively disable traumatic memories in victims of post-traumatic stress disorder; by 2039 everyone will have an Interlocutor, a software agent that answers the phone for us, manages our investments, and arranges dates for us… we could do this all day, and Kurzweil probably does). The winner is the first person to sneak in a prediction of something that has in fact happened already.

But beneath the froth is a sharp and original mind which it would be all too easy to underestimate. Why did Google want him? The Guardian frames the shopping spree as being about bringing together the best experts and the colossal data resources to which Google has access. A plausible guess would be that Google wants to improve its core product dramatically. At the moment Google answers questions by trying to provide a page from the web where some human being has already given the answer; perhaps the new goal is technology that understands the question so well that it can put together its own answer, gathering and shaping selected resources in very much the way a human researcher working on a bespoke project might do.

But perhaps it goes a little further: perhaps they hope to produce something that will interact with humans in a human-like way.  A piece of software like that might well be taken to have passed the Turing test, which in turn might be taken to show that it was, to all intents and purposes, a conscious entity. Of course, if it wasn’t conscious, that might be a disastrous outcome; the nightmare scenario feared by some in which our mistake causes us to nonsensically award the software human rights, and/or  to feel happier about denying them to human beings.

It’s not very likely that the hypothetical software (and we must remember that this is the merest speculation) would have even the most minimal forms of consciousness. We might take the analogy of Google Translate; a hugely successful piece of kit, but one that produces its translations with no actual understanding of the texts or even the languages involved. Although highly sophisticated, it is in essence a ‘brute force’ solution; what makes it work is the massive power behind it and the massive corpus of texts it has access to.  It seems quite possible that with enough resources we might now be able to produce a credible brute force winner of the Turing Test: no attempt to fathom the meanings or to introduce counterparts of human thought, just a massive repertoire of canned responses, so vast that it gives the impression of fully human-style interaction. Could it be that Google is assembling a team to carry out such a project?

Well, it could be. However, it could also be that cracking true thought is actually on the menu. Vaughan Bell suggests that the folks recruited by Google are honest machine learning types with no ambitions in the direction of strong AI. Yet, he points out, there are also names associated with the trendy topic of deep learning. The neural networks (but y’know, deeper) which deep learning uses just might be candidates for modelling human neuron-style cognition. Unfortunately it seems quite possible that if consciousness were created by deep learning methods, nobody would be completely sure how it worked or whether it was real consciousness or not. This is a lamentable outcome: it’s bad enough to have robots that naive users think are people; having robots and genuinely not knowing whether they’re people or not would be deeply problematic.

Probably nothing like that will happen: maybe nothing will happen. The Guardian piece suggests Kurzweil is a bit of an outsider: I don’t know about that.  Making extravagantly optimistic predictions while only actually delivering much more modest incremental gains? He sounds like the personification of the AI business over the years.