The original is still somewhat acceptable as a Fermi estimation and argument, as AlphaGo still uses more energy (3600 times more) with this conservative estimate.
I'm not sure why having a "fair" competition matters. The point of technological progress is to get better outcomes, not to get better outcomes using massively constrained hardware. Of course the latter is useful, but it's totally arbitrary to just pick energy usage as the one constraint to put on computers, there are hundreds of other ways in which they're advantaged. The important thing is that it can be done at all.
There are VERY few situations where I care even remotely how energy efficient someone's thinking is. The benefits of better thinking are nearly always far more valuable than the energy inputs. We are frequently limited by the fact that no more intelligence can be applied to the problem no matter what we do. In many other cases we are limited by cost -- we would happily hire a team of three planners or one expensive genius, whichever was smart enough to lead our project at the cheaper price. But even in extreme cases like manned spaceflight, I've never heard of hiring someone dumber because at least they ate less.
Energy efficiency does matter for small-scale stuff like mobile devices, and for medium-to-large-scale stuff like servers (unless money is no object). But yes, it's not really all that important for something like AI research. Especially considering that at the rate AlphaGo has been improving,* by some time in the next year or two it should be able to beat Sedol running on a smartphone.
*And yes, there's no guarantee it will continue improving at the same rate.
That's not the point at all. The point is that if you decide tomorrow to run thousands of AIs like AlphaGo to replace humans to do several stuff, does that scale well? Well, if it consumes as much power as AlphaGo, it certainly does not. The point about having something running on low watts is critical for scalability in the longer term, because no matter how you look at it, energy production is finite (you can increase it over time but there are clear limits anyway).
Machine learning is compatible with low energy applications. There is a body of research in slimming down heavy neural nets to run on cell phones with very little accuracy loss. AlphaGo uses so much power because it additionally uses a brute force approach to improve its results.
IMHO the whole humanity won't contribute much to the challenge. You take the best 5 pros and that's pretty much the best team you can pit against the AI. Ten pros would probably need too much time to discuss and coordinate. A striking analogy with software dev teams :-)
The human brain is in this situation much closer to an ASIC than a CPU. An average human can't play go on this level of skill – you need to use 20 years to turn your brain into the right kind of ASIC to play go.
Also comparing the human brain and cpus in general, it should be noted that we are just good at different things. When it comes to basic arithmetic, human brains are much less energy-efficient than cpus.
Calling a human brain an ASIC seems ridiculous to me. Lee Sedol, in addition to being a marvellous Go player, is also a typical human being. He can speak a language, read and write, probably drive a car, take care of his wife and child. AlphaGo is incapable of all those things. If you ask me, it's closer to an ASIC than a human is.
The AlphaGo team tweeted that the single node version wins 25% of the time against the cluster version.
The bulk of the compute is in training the model... I would bet that a cell phone AlphaGo is <200 ELO (or whatever passes in the Go world) weaker than the massively distributed version--good enough to be competitive with Lee Sedol.
If it works for chess, it'll work for Go. Chess has lots of games that you can learn from, Komodo wins any grandmaster or draws.
The problem with Go was lack of evaluation function that would guide the policy. So it had to be learned simultaneously.
You can leave AlphaGo to play a billion games and then learn a policy that requires little to no search but has almost perfect evaluation (local optimality of minimizing future regret).
Same positional play is exhibited by Komodo, and it requires not that much of depth searching, while currently AlphaGo rolls out a whole game for every move.
Give it a few years and you can draw your conclusions, guaranteed.
60 years ago: "The ENIAC uses so much more energy and takes up more space than a human to multiply numbers, I want to see a calculator multiply faster than a human running on a 5V watch battery before I draw conclusions"
> In the one corner: Human, all of 175 pounds of extremely well trained runner. And in the other corner, a Formula 1 racecar with a remote control running down a straight track.
>
> Nonsense, you’d say, that’s not a fair comparison
No, I (and I think most sensible people) would say that's pointless – we know who's going to win. If you have a sufficiently smooth surface or it's worth creating one, cars are way better at their specific task than humans are. Everybody knows that and nobody uses humans for transportation or walks when they care about getting there fast.
Why is everybody so scared that computers might now “be better” than humans?
To say they are categorically better implies better at all things. Now a computer eclipses any human in arithmetic, memorization, etc, but even thousands of cores cannot make it do simple things even toddlers can do.
On top of that, when trains (and probably cars too) were first introduced, there were races between trains and horses to see which was faster. Nowadays that's pointless for anything other than short distances, but early on it was a good measure of the technology, even if there are "obvious" advantages towards the train/car/AlphaGo.
I'm not scared at all. I'm just pointing out that there is another hurdle that would be a major achievement to cross, without belittling at all what it took to achieve this victory. It is one of the most interesting results for many years simply because almost all the experts agreed this was at least a decade away.
> without belittling at all what it took to achieve this victory
That's not really what it comes across as. Most posts like that reek of moving the goalpost.
Also the efficiency goalpost is probably the easier problem. Advancement in processor designs, microarchitecture, and basic code optimization will get us there, not necessarily changing the techniques of AlphaGo.
* 60 years ago ENIAC was room-sized and slow, now salesmen hand out branded solar calculators for free
* As the article mentioned, deepblue was a massive computer with custom hardware, now your <1W phone can run stockfish. The achievement wasn't by the chess team, it was by Intel, ARM, compiler developers, etc.
Shifting goalposts. And humans cost much more energy than the watts contained in their food. Humans take 20 years to "train" and need lots of water, electricity, food, clothes, transportation and teaching time.
If they were to use only the policy net part of AlphaGo they would need a few milliseconds per move and only one computer to run it, achieving the level of 1p. Also, if DeepMind used specialized Go hardware maybe they could reduce the power usage a lot, but they didn't optimize for that.
For example, it would be interesting to calculate how much power is used in image recognition, man vs machine. The brain uses 20-40W and a cell phone only 5W, and we know cell phones can do image recognition, so, probably the computer uses less than the human given that human reaction time is 200ms and the computer could classify much more than 5 images per second, so it uses less time and consequently less energy per image.
>The brain uses 20-40W and a cell phone only 5W, and we know cell phones can do image recognition, so, probably the computer uses less than the human given that human reaction time is 200ms and the computer could classify much more than 5 images per second, so it uses less time and consequently less energy per image.
I'm all for praising the enormous developments being made in modern computing, but this is absolutely needless exaggeration. Computers are far, far, far weaker at image recognition than even an 8 year old child. The amount of things humans can continually process, using a fraction of the brain (the brain is constantly doing visual processing without shutting the other centers down), whereas a computer can barely do simple tasks and takes longer to do that than "200ms", as you claim.
The brain is a mind boggling machine. Our computers are getting pretty kickass, but there is still a long long way to go.
Just because we're doing things that way doesn't mean there's a requirement to do so.
I read that when the British government wanted people to translate Japanese for them in WWII, they went to their universities and heard, from the relevant departments, that becoming sufficiently conversant in Japanese was a painstaking, multi-year process. Someone else heard about this, thought "pffft, I could train people to be conversant in Japanese in six months", and set up a school. He was right, the university departments were wrong.
You see so many (justified!) complaints that modern schooling is composed mostly of wasted time. Humans don't take 20 years to train.
Train for a specific task? No. But train to be able to acquire new skills, yes. For example, I have explained the basics of Go to my 5yr-old, but he will need years to be able to develop sufficient abstract thinking, patience, etc. to actually play.
Schooling is not primarily intended for academic training, at least until post-secondary levels are reached. Instead, it's primarily about socializing, instilling rule-following, reinforcing social roles, and so on.
I agree it would be a good challenge, but I disagree that it would be "fairer" and "much more important" than AlphaGo itself.
The fact that DeepMind accomplished this feat, period, is what's most important. The fact that it runs on Google Cloud Platform, not custom-built hardware (à la Deep Blue), is almost as important. When it inevitably shrinks down, that would be icing on the cake.
I mean, you could keep counting all the energy used in the causal chain as far back as you want. After all, AlphaGo was created by humans whose brains used a heck of a lot of energy.
I still think that looking at marginal energy cost while playing the game is an interesting approach.
AlphaGo is only two years old and is one of a handful of computer AI programs, but humans have had millions of years of evolution. So it's not fair to compare. Computers have evolved too, maybe even more dramatically than biological life, to lower and lower levels of power usage.
In Chess no top player can beat his/her cell phone. This milestone will be eventually achieved in Go too.
How much energy do you think went into "training" alphago? Not just electricity and server time to train it's neural networks, but the development team too.
Do you think a thousand amateurs working together could be Lee Sedol? I'm inclined to believe that having 1000 of them would only make an impossible feat harder. Strategy quality doesn't, in fact, scale with the number of individual actors creating the strategy.
You could also look at cost. How much would it cost to rent the alphago processing power from EC2, and how much does it cost to get a top player to agree to play you? (The actual match here doesn't count because of publicity, but I imagine general costs should be easily calculated.)
It's not simply an algorithm. It as learning computer. It remembers and learns from previous matches. He can probably play more matches against AlphaGo if he wants to but the computer will keep getting better and probably faster than he is.
I don't understand why a Joule-for-Joule "fair" competition will seem like a more important achievement. Computers can trade efficiency for raw performance, humans can't -- it's an unfair advantage. How the AlphaGo team balances those two over time will change, but for now I'm delighted to see what happens when they optimize for something that is beyond human potential.
Agreed. The amount of energy it takes to do this will soon fit in my phone easily without having an impact on my battery life (as is the case with chess).
Why do we count only the amount of energy the human uses during the match? After the match we can turn the machine off, making it use no energy. Humans need to spend 50+% of their time doing stuff that isn't even related to the activities we're measuring (eat, sleep, do laundry, drink wine with friends), and 99% of the rest on 'training'.
I fail to see why energy-parity comparisons are 'more fair' or 'better'. We don't have soccer matches for people who eat nothing but two sandwiches a day, do we? Has a marathon winner cheated if he ate a few bananas during the race and the runner up hasn't? Of course not (well, 'of course' to me, it seems that the OP might disagree?).
I guess we could move on to comparing sporting wins by people who used steroids, but that would just stray from the actual point.
That said, I don't think it's the same. Vehicles and fighters are classified in various ways (power/weight, but also experience, e.g. pro and amateur classes) because otherwise the competition becomes 'who has the best engineering team' and not 'who drives best'; or 'who has most fighting skills' and not 'who can overpower his opponent through sheer force/mass'. It's purely to keep it interesting and/or exciting.
In this case, I interpreted the argument made in the OP as 'it only becomes an achievement when the computer constrains itself to the boundaries the human has by its nature'. I don't see the point of that. An F1 car is unarguably faster than 50cc kart. If we talk about 'who can drive the fastest', it doesn't make sense to say 'it's not a 'fair' comparison'. It is, and the F1 car is faster, period.
The nature of the past match was 'who is the best Go player, no holds barred'. Call it the UFC 1 of Go. Back in 1994 in UFC 1 there were no weight classes, no restrictions on techniques - the question was 'who is the best fighter'. It's not like after that event the sumo guy said 'yeah well I can't move as fast because I'm 200kg, so we'll only know who is the best fighter after a jiu jitsu guy fights me but doesn't move faster than I do'.
I guess there is more nuance to the debate than I originally thought. I still think it doesn't make sense to impose artificial limits. We might have tournaments for humans and tournaments for machines in the future, just like we have nascar and F1. But there is no mistake who is the best player.
>I interpreted the argument made in the OP as 'it only becomes an achievement when the computer constrains itself to the boundaries the human has by its nature'
I don't think this is a particularly accurate interpretation considering the author says:
>Now, not to diminish the achievement of the AlphaGo team, what they have done is nothing short of incredible
I think a more charitable interpretation can be found by looking at the closing paragraph which starts with this line:
>So now the interesting question (to me at least) is: How long before a computer will beat the human Go world champion using no more power than the human.
I think too much examination may get to the point where the competition analogy isn't helpful. The point is, OK, that is a major achievement, but now can we make it work using standard consumer hardware? To bring another analogy into the mix, the current situation is as though we've gotten a team of members of our chess club who get to all consult each other during the match and don't have to follow the clock to beat a chess grandmaster -- a major achievement, to be sure, but we can hardly call ourselves his equal.
Also, it's fairly easy to replicate the Go-playing machine while good human Go players are very rare (and how about counting the resources needed to produce an environment where such a player can develop? Then again we could count the resources going into developing computer hardware...), and they cannot be predictably "produced."
Describing this as "a computer beats a human for the first time but using a lot of energy" somewhat misrepresents it because a Go bot running on a mobile device will wipe the floor with the vast majority of humans - and most amateur players, because there's a power law distribution with these things and so most players who're any good, including bots, are better than the majority of players (while at the same time being not even remotely interesting to play with for a very large number of stronger players.)
The point is that any mobile device can be "trained" with a few clicks to wipe the floor with most humans, for whom to catch up with the bot would require a lot of energy (and perhaps isn't always possible.) Does it make phones smarter the people? I dunno, but it certainly shows that these comparisons are never "fair" and the only real question is what practical implications of "intelligence" you care about, then you can evaluate things meaningfully.
We could consider lifetime costs. A human requires other humans to build, takes 16+ years to grow to reasonable capacity, can't work 24/7 or even a consistent 40 hours a week...
That's why I fully expect supermarkets to have drastically reduced labor costs in a few years, as shelved goods will be restocked by micro-forklift robots, checkouts will be automated and the human staff is reduced to two or three supervisory and customer-relations positions.
Except AlphaGo isn't eating 2 fingers of banana; it's eating enough quantity to starve a whole suburb. See the OP as a finger pointing direction of what the next exciting challenge is instead of as a criticism of the success of DeepMind/AlphaGo.
Another comparison would be how would the Formula 1 car do in terms of sourcing its own energy to keep going. Humans - in our primitive state were adapted for 'persistence hunting' - we ran in packs and drove animals into hyperthermia.
The Formula 1 car would do quite well for the first few hundred miles. But you would eventually get an Aesopian 'tortoise and hare' effect.
In the real world the OP's point is a good one. Lee SeDol could outlast AlphaGo by making it use too much in the way of resources (energy, human work, maintenance) untill he could get close enough to fell the beast with a stone arrow or hardened wood spear...
F1 cars are good at causing humans to tend to their needs, though! Much better than most humans. The average F1 car gets far more resources than, for example, I do.
Here's a fun idea. Chess is thoroughly lost to the computers at this point, but what if we add in an element that would give the humans an advantage? There's a real actual sport called Chess boxing. The competitors spend three minutes playing chess, then three minutes pounding the crap out of each other, alternating back and forth. Let's see how well a modern computer can do at this game!
I'm sure that without substantial modification that would violate some of the rules of the game. Chess boxing also has weight classes, so while it might be able to win a heavyweight match, other weight classes would be much more interesting.
To keep things fair, I think we should require the computer to view and manipulate the physical chess board and pieces. None of this nonsense of feeding it moves through a keyboard and taking instructions off a screen.
Why create a bunch of rules to keep things fair? Man vs machine is never going to be fair once we have the machine to do it. Before deep blue or alphaGo, yea it'll be a fun event. Once the machine exists, we check off another box in the list of things that machines can now do, and we move on with our lives.
Though on the topic of the movement aspect of boxing, machines are still pretty bad at keeping their balance and moving around the world, even without limitations.
Don't create new rules, just keep existing ones. The rules I found on worldchessboxing.com (every time I see the phrase "chess boxing" it makes me laugh) say that the chess clock must be pressed with the same hand that moves the piece, so apparently the computer is going to need hands (well, one hand). They also say that it's the player's responsibility to press their own clock between moves, so that rules out any outside help for moving pieces.
I note that the players are allowed to bring water to the chess table. I'm not sure if there's a rule against pouring it on your opponent. The computer might need to take this into account.
I think a <200lb bipedal 2-arm robot which could box would be a marvel to see, but if there aren't any rules which say how many arms and legs the combatant has to have, the robot is gonna have wheels, most of its mass in the lower body so as to make it un-topplable, be made of some lightweight metal, and have an arm that plays chess and a boxing glove on an actuator that can break concrete.
And on the point of existing rules, the existing rules aren't enough for this type of competition. First of all the weight class: are robots men or women? It says age 17+, how does that affect the robot? There's no rule against pouring water on the robot, but there also isn't a rule against releasing nerve gas. My robot is a pourous metal box with a canister of gas which renders everyone unconscious, do I win by K/O?
Also, they already have battle robots, and they look nothing like conventional humans (one common design is a flat panel that slides under the opponent and jettisons them)
My main complaint with this line of thought is this. It is claimed that it is not fair due to energy usage comparisons, but if you asked people before AlphaGo if it was a "fair" competition to pit a Go AI versus a world class professional, the answer would be hilariously no, because the world class professional would destroy the AI. I imagine telling them that you'll use more energy than them wouldn't change the consensus either. But now that a computer using so much energy did beat a world class professional, now this same scenario is framed as unfair for the professional.
There's plenty of back and forth arguments that can be made as to whether the energy argument is a true test of fairness, but I don't think it matters if it's a good yardstick for fairness or not. It was an amazing feat to create an AI of any size that could beat a professional.
The post starts with an analogy of a race between a runner and a race car. It says we'd call this unfair because the two are clearly not comparable.
But I think it gets the reasons wrong. It says we'd call it unfair because the race car uses so much more energy. I think we'd call it unfair because the outcome is such a foregone conclusion. We know the race car is going to be way faster, so the race is pointless. It's a foregone conclusion, so why even bother unless you do something to make it more even?
Go back to the very early days of automobiles, when they were so slow that a human runner would be faster. Then a new one comes along that can beat a human. The first car faster than a person!
Would you be amazed that machine has beaten man? Or would you say that it somehow doesn't count, because machine used more fuel than man? I'm pretty sure most people would not go for the second one.
I find people using various tricks to somehow console themselves of the fact that a machine beat a human to be quite funny.
What is it exactly that they're trying to protect? Perhaps at the core is a fear that if someone or some thing does something better than me than I'm not worth as much, or more generally, if "they" (the machines) do things better than "us" then we're not worth as much, and what wouldn't we do to feel worthy... but I feel this whole nonsense stems from a belief that anything in existence could really be unworthy.
I remember a short passage, from one of Raymond Smullyan's books noting two different responses from humans the day it is discovered that it is possible to make machines that are indistinguishable from humans. The first says glumly: "So, we are just machines?" The other says with joy: "I didn't know machines could be so wonderful".
It's not a contest of worthiness - it's reality beautifully playing out and there are no unworthy players.
> I find people using various tricks to somehow console themselves of the fact that a machine beat a human to be quite funny.
I find it quite funny that you'd take what I wrote and that you'd manage to wrangle from it that we need to be consoled. I simply observed that there is another challenge, one that in chess has since the first computer beat a world champion been more than met and that in Go there is still some distance left to cover.
Put another way, if the size and energy consumption would not matter do you think that there would have been improvement in the Chess programs post the point where they could beat any chess player in the world? Clearly the Chess programming community thought otherwise and we've seen a major improvement in algorithms and this resulted in a huge decrease in the required computing horsepower. That's smarter programming, not using larger computers to achieve the win through brute force and to me - feel free to disagree - that's an interesting prospect.
Ok, I made an unnecessary leap from knowing that many people do feel threatened by the increasing problem solving abilities of computers to assuming you are one of them. so I stand corrected, and in light of your response understand that my sermon was misplaced in this context.
I don't disagree with you at all. In fact, I think probably we're headed for the same improvements over time with machines playing Go as was with machines playing Chess. I think that in addition to better algorithms, Chess had the advantage of having Moore's law being in full swing over these years, which will probably slow down, but there's always something new and unexpected coming up, so beyond multiple processors and GPU's with lower energy consumption I'm sure there will be even further improvements in the machine learning methods and other things that I can't yet imagine.
This isn't a better view though. The number of Watts is not important. Biology is efficient in energy use, so what? If you tax a human brain it doesn't especially consume vast amounts more energy. I don't work up a sweat thinking. What's more interesting is if the computational "power" has parity. Deep Blue wasn't as computationally powerful as a human brain. It was specialised which is why it's not as interesting a victory as AlphaGo which is based on a more general principal.
That's a very bold statement to make coming from the perspective of an industry that does just about everything it can to save energy. Better programming means lower energy consumption, to achieve a win like this on 1% of the energy budget would be a major game changer (no pun intended).
AlphaGo is still using brute force quite a bit, the stage is set for a much improved batch of software that will focus less on brute force as a main strategy.
This is exactly the question I was asking myself this morning! I find the parallels to robotics interesting too: whenever another creepy beast lurches out of Boston Dynamics I wonder how such machines will function 50 or 100 year in the future when the world is operating under a different energy cost equation, when the fruits of millions of years of fossil fuel deposit have been spanked.
Edit: I see lots of "no fair!" comments here. I think people are failing to grasp how supremely efficient biological systems are at dealing with exactly the sort of messy/noisy/lossy reasoning required for real-world problem solving, the sort of reasoning I see linked (incorrectly IMHO) by various media to the advancement demonstrated by this hyper-specialized Go playing machine.
Edit2: (I removed that "little to show for it" bit before I saw your comment because it was leading to a digression I realised I would need to defend which seemed unwarranted in this context. Sorry about that!)
Those fossil fuels allowed us to rapidly bootstrap ourselves through the industrial revolution, and to get to the point where renewable and carbon-neutral energy isn't just feasible, but already being used in many places.
It is hypocritical, but that doesn't mean it's not a good idea. That's also not to say it isn't necessarily unfair. I tend to see it like progression in a game's tech tree, in some sense the prior steps (oil) were needed to get to a point where we can actually create energy in a clean and renewable way. In the future, there will likely be a tipping point where the energy created from clean and renewable sources can be used to help create new sources, but it likely wouldn't have been possible without the oil stepping stone. In an ideal and non-hypocritical scenario, the advancements in energy production would be shared to avoid the dirty oil intermediary step, but while there is still scarcity and separate national entities, there will still be hoarding and greed.
A great implementation of this is type of thinking is Morocco's solar plant in the Sahara [0]. Though I don't think this would have happened if it wasn't also thought to be a good investment, and urging a bad investment that you're not willing to participate in would definitely be hypocritical.
Big problem in this way of looking at this: energy is measured in Joules, not in Watts.
So this is not a fair comparison. Yes Lee was pretty cheap energy-wise to "run" during the game itself, but needed 150W or so constantly for 33 years to get trained to be as good as he is (brain is cheap, but obviously you have to keep the body alive too). Plus the massive support structure needed to raise him, teach him, get him interested in Go, provide opponents to train against. As you can't have a grandmaster. You need 10 grandmasters, 1000 really good players, 1000000 good players, and so on to train someone to get to this level.
AlphaGo needed 1MW, probably about the same for training. If AlphaGo got trained in about a week, I bet the energy difference would favor AlphaGo, depending on what you count you could make AlphaGo or Lee Sedol win the comparison, so really, it's a tossup.
And let's compare it to chess. Deep Blue used probably a similar amount of power in it's entire lifetime as Gary Kasparov will need for his lifetime. Maybe a factor 10 difference, but no more. However, a current chess computer uses less than a billionth the amount of power for learning & playing chess than Gary Kasparov needs to live and learn to play chess at his level.
112 comments
[ 2.8 ms ] story [ 157 ms ] threadThe TDP of the 24 core Xeon systems that were most likely used is 6.8W per core.
*And yes, there's no guarantee it will continue improving at the same rate.
If no - computers still win because they scale better (play better as a team).
Given how well collective intelligence worked in that game, it'd be pretty interesting to try it against AlphaGo.
Also comparing the human brain and cpus in general, it should be noted that we are just good at different things. When it comes to basic arithmetic, human brains are much less energy-efficient than cpus.
One interesting thing I found by Googling FPGA (which is the easisest way to build a new chip) and neural networks, Microsoft is building one which requires considerabley less power: https://gigaom.com/2015/02/23/microsoft-is-building-fast-low...
The bulk of the compute is in training the model... I would bet that a cell phone AlphaGo is <200 ELO (or whatever passes in the Go world) weaker than the massively distributed version--good enough to be competitive with Lee Sedol.
The problem with Go was lack of evaluation function that would guide the policy. So it had to be learned simultaneously.
You can leave AlphaGo to play a billion games and then learn a policy that requires little to no search but has almost perfect evaluation (local optimality of minimizing future regret).
Same positional play is exhibited by Komodo, and it requires not that much of depth searching, while currently AlphaGo rolls out a whole game for every move.
60 years ago: "The ENIAC uses so much more energy and takes up more space than a human to multiply numbers, I want to see a calculator multiply faster than a human running on a 5V watch battery before I draw conclusions"
No, I (and I think most sensible people) would say that's pointless – we know who's going to win. If you have a sufficiently smooth surface or it's worth creating one, cars are way better at their specific task than humans are. Everybody knows that and nobody uses humans for transportation or walks when they care about getting there fast.
Why is everybody so scared that computers might now “be better” than humans?
That's not really what it comes across as. Most posts like that reek of moving the goalpost.
Also the efficiency goalpost is probably the easier problem. Advancement in processor designs, microarchitecture, and basic code optimization will get us there, not necessarily changing the techniques of AlphaGo.
* 60 years ago ENIAC was room-sized and slow, now salesmen hand out branded solar calculators for free * As the article mentioned, deepblue was a massive computer with custom hardware, now your <1W phone can run stockfish. The achievement wasn't by the chess team, it was by Intel, ARM, compiler developers, etc.
If they were to use only the policy net part of AlphaGo they would need a few milliseconds per move and only one computer to run it, achieving the level of 1p. Also, if DeepMind used specialized Go hardware maybe they could reduce the power usage a lot, but they didn't optimize for that.
For example, it would be interesting to calculate how much power is used in image recognition, man vs machine. The brain uses 20-40W and a cell phone only 5W, and we know cell phones can do image recognition, so, probably the computer uses less than the human given that human reaction time is 200ms and the computer could classify much more than 5 images per second, so it uses less time and consequently less energy per image.
I'm all for praising the enormous developments being made in modern computing, but this is absolutely needless exaggeration. Computers are far, far, far weaker at image recognition than even an 8 year old child. The amount of things humans can continually process, using a fraction of the brain (the brain is constantly doing visual processing without shutting the other centers down), whereas a computer can barely do simple tasks and takes longer to do that than "200ms", as you claim.
The brain is a mind boggling machine. Our computers are getting pretty kickass, but there is still a long long way to go.
It still uses a lot of hardware, but not hardware built for the sole purpose of playing Go.
Just because we're doing things that way doesn't mean there's a requirement to do so.
I read that when the British government wanted people to translate Japanese for them in WWII, they went to their universities and heard, from the relevant departments, that becoming sufficiently conversant in Japanese was a painstaking, multi-year process. Someone else heard about this, thought "pffft, I could train people to be conversant in Japanese in six months", and set up a school. He was right, the university departments were wrong.
You see so many (justified!) complaints that modern schooling is composed mostly of wasted time. Humans don't take 20 years to train.
Schooling is not primarily intended for academic training, at least until post-secondary levels are reached. Instead, it's primarily about socializing, instilling rule-following, reinforcing social roles, and so on.
No, new challenge.
The fact that DeepMind accomplished this feat, period, is what's most important. The fact that it runs on Google Cloud Platform, not custom-built hardware (à la Deep Blue), is almost as important. When it inevitably shrinks down, that would be icing on the cake.
I still think that looking at marginal energy cost while playing the game is an interesting approach.
In Chess no top player can beat his/her cell phone. This milestone will be eventually achieved in Go too.
In the world of Chess it took until 1996 before a computer won against the then reigning world champion, Gary Kasparov in a series of 6 matches.
1) His name is Garry Kasparov
2) Kasparov won the 1996 series against Deep Blue 4-2, it was in 1997 that Kasparov lost 2.5-3.5
Starcraft, anyone?
I fail to see why energy-parity comparisons are 'more fair' or 'better'. We don't have soccer matches for people who eat nothing but two sandwiches a day, do we? Has a marathon winner cheated if he ate a few bananas during the race and the runner up hasn't? Of course not (well, 'of course' to me, it seems that the OP might disagree?).
I guess we could move on to comparing sporting wins by people who used steroids, but that would just stray from the actual point.
That said, I don't think it's the same. Vehicles and fighters are classified in various ways (power/weight, but also experience, e.g. pro and amateur classes) because otherwise the competition becomes 'who has the best engineering team' and not 'who drives best'; or 'who has most fighting skills' and not 'who can overpower his opponent through sheer force/mass'. It's purely to keep it interesting and/or exciting.
In this case, I interpreted the argument made in the OP as 'it only becomes an achievement when the computer constrains itself to the boundaries the human has by its nature'. I don't see the point of that. An F1 car is unarguably faster than 50cc kart. If we talk about 'who can drive the fastest', it doesn't make sense to say 'it's not a 'fair' comparison'. It is, and the F1 car is faster, period.
The nature of the past match was 'who is the best Go player, no holds barred'. Call it the UFC 1 of Go. Back in 1994 in UFC 1 there were no weight classes, no restrictions on techniques - the question was 'who is the best fighter'. It's not like after that event the sumo guy said 'yeah well I can't move as fast because I'm 200kg, so we'll only know who is the best fighter after a jiu jitsu guy fights me but doesn't move faster than I do'.
I guess there is more nuance to the debate than I originally thought. I still think it doesn't make sense to impose artificial limits. We might have tournaments for humans and tournaments for machines in the future, just like we have nascar and F1. But there is no mistake who is the best player.
I don't think this is a particularly accurate interpretation considering the author says:
>Now, not to diminish the achievement of the AlphaGo team, what they have done is nothing short of incredible
I think a more charitable interpretation can be found by looking at the closing paragraph which starts with this line:
>So now the interesting question (to me at least) is: How long before a computer will beat the human Go world champion using no more power than the human.
Describing this as "a computer beats a human for the first time but using a lot of energy" somewhat misrepresents it because a Go bot running on a mobile device will wipe the floor with the vast majority of humans - and most amateur players, because there's a power law distribution with these things and so most players who're any good, including bots, are better than the majority of players (while at the same time being not even remotely interesting to play with for a very large number of stronger players.)
The point is that any mobile device can be "trained" with a few clicks to wipe the floor with most humans, for whom to catch up with the bot would require a lot of energy (and perhaps isn't always possible.) Does it make phones smarter the people? I dunno, but it certainly shows that these comparisons are never "fair" and the only real question is what practical implications of "intelligence" you care about, then you can evaluate things meaningfully.
That's why I fully expect supermarkets to have drastically reduced labor costs in a few years, as shelved goods will be restocked by micro-forklift robots, checkouts will be automated and the human staff is reduced to two or three supervisory and customer-relations positions.
The Formula 1 car would do quite well for the first few hundred miles. But you would eventually get an Aesopian 'tortoise and hare' effect.
In the real world the OP's point is a good one. Lee SeDol could outlast AlphaGo by making it use too much in the way of resources (energy, human work, maintenance) untill he could get close enough to fell the beast with a stone arrow or hardened wood spear...
https://en.wikipedia.org/wiki/Endurance_running_hypothesis
Here's a fun idea. Chess is thoroughly lost to the computers at this point, but what if we add in an element that would give the humans an advantage? There's a real actual sport called Chess boxing. The competitors spend three minutes playing chess, then three minutes pounding the crap out of each other, alternating back and forth. Let's see how well a modern computer can do at this game!
I suggested a combination of the two and he told me about Chess boxing:
https://en.wikipedia.org/wiki/Chess_boxing
To keep things fair, I think we should require the computer to view and manipulate the physical chess board and pieces. None of this nonsense of feeding it moves through a keyboard and taking instructions off a screen.
Though on the topic of the movement aspect of boxing, machines are still pretty bad at keeping their balance and moving around the world, even without limitations.
I note that the players are allowed to bring water to the chess table. I'm not sure if there's a rule against pouring it on your opponent. The computer might need to take this into account.
And on the point of existing rules, the existing rules aren't enough for this type of competition. First of all the weight class: are robots men or women? It says age 17+, how does that affect the robot? There's no rule against pouring water on the robot, but there also isn't a rule against releasing nerve gas. My robot is a pourous metal box with a canister of gas which renders everyone unconscious, do I win by K/O?
Also, they already have battle robots, and they look nothing like conventional humans (one common design is a flat panel that slides under the opponent and jettisons them)
As for reaction speed of man (100+ms vs machine), I'll just link this: https://www.youtube.com/watch?v=3nxjjztQKtY
This tournament is sounding better and better all the time!
There's plenty of back and forth arguments that can be made as to whether the energy argument is a true test of fairness, but I don't think it matters if it's a good yardstick for fairness or not. It was an amazing feat to create an AI of any size that could beat a professional.
But I think it gets the reasons wrong. It says we'd call it unfair because the race car uses so much more energy. I think we'd call it unfair because the outcome is such a foregone conclusion. We know the race car is going to be way faster, so the race is pointless. It's a foregone conclusion, so why even bother unless you do something to make it more even?
Go back to the very early days of automobiles, when they were so slow that a human runner would be faster. Then a new one comes along that can beat a human. The first car faster than a person!
Would you be amazed that machine has beaten man? Or would you say that it somehow doesn't count, because machine used more fuel than man? I'm pretty sure most people would not go for the second one.
What is it exactly that they're trying to protect? Perhaps at the core is a fear that if someone or some thing does something better than me than I'm not worth as much, or more generally, if "they" (the machines) do things better than "us" then we're not worth as much, and what wouldn't we do to feel worthy... but I feel this whole nonsense stems from a belief that anything in existence could really be unworthy. I remember a short passage, from one of Raymond Smullyan's books noting two different responses from humans the day it is discovered that it is possible to make machines that are indistinguishable from humans. The first says glumly: "So, we are just machines?" The other says with joy: "I didn't know machines could be so wonderful". It's not a contest of worthiness - it's reality beautifully playing out and there are no unworthy players.
I find it quite funny that you'd take what I wrote and that you'd manage to wrangle from it that we need to be consoled. I simply observed that there is another challenge, one that in chess has since the first computer beat a world champion been more than met and that in Go there is still some distance left to cover.
Put another way, if the size and energy consumption would not matter do you think that there would have been improvement in the Chess programs post the point where they could beat any chess player in the world? Clearly the Chess programming community thought otherwise and we've seen a major improvement in algorithms and this resulted in a huge decrease in the required computing horsepower. That's smarter programming, not using larger computers to achieve the win through brute force and to me - feel free to disagree - that's an interesting prospect.
I don't disagree with you at all. In fact, I think probably we're headed for the same improvements over time with machines playing Go as was with machines playing Chess. I think that in addition to better algorithms, Chess had the advantage of having Moore's law being in full swing over these years, which will probably slow down, but there's always something new and unexpected coming up, so beyond multiple processors and GPU's with lower energy consumption I'm sure there will be even further improvements in the machine learning methods and other things that I can't yet imagine.
That's a very bold statement to make coming from the perspective of an industry that does just about everything it can to save energy. Better programming means lower energy consumption, to achieve a win like this on 1% of the energy budget would be a major game changer (no pun intended).
AlphaGo is still using brute force quite a bit, the stage is set for a much improved batch of software that will focus less on brute force as a main strategy.
Edit: I see lots of "no fair!" comments here. I think people are failing to grasp how supremely efficient biological systems are at dealing with exactly the sort of messy/noisy/lossy reasoning required for real-world problem solving, the sort of reasoning I see linked (incorrectly IMHO) by various media to the advancement demonstrated by this hyper-specialized Go playing machine.
Edit2: (I removed that "little to show for it" bit before I saw your comment because it was leading to a digression I realised I would need to defend which seemed unwarranted in this context. Sorry about that!)
Those fossil fuels allowed us to rapidly bootstrap ourselves through the industrial revolution, and to get to the point where renewable and carbon-neutral energy isn't just feasible, but already being used in many places.
I'd say we've got quite a bit to show for it.
A great implementation of this is type of thinking is Morocco's solar plant in the Sahara [0]. Though I don't think this would have happened if it wasn't also thought to be a good investment, and urging a bad investment that you're not willing to participate in would definitely be hypocritical.
[0]:http://www.npr.org/sections/thetwo-way/2016/02/04/465568055/...
So this is not a fair comparison. Yes Lee was pretty cheap energy-wise to "run" during the game itself, but needed 150W or so constantly for 33 years to get trained to be as good as he is (brain is cheap, but obviously you have to keep the body alive too). Plus the massive support structure needed to raise him, teach him, get him interested in Go, provide opponents to train against. As you can't have a grandmaster. You need 10 grandmasters, 1000 really good players, 1000000 good players, and so on to train someone to get to this level.
AlphaGo needed 1MW, probably about the same for training. If AlphaGo got trained in about a week, I bet the energy difference would favor AlphaGo, depending on what you count you could make AlphaGo or Lee Sedol win the comparison, so really, it's a tossup.
And let's compare it to chess. Deep Blue used probably a similar amount of power in it's entire lifetime as Gary Kasparov will need for his lifetime. Maybe a factor 10 difference, but no more. However, a current chess computer uses less than a billionth the amount of power for learning & playing chess than Gary Kasparov needs to live and learn to play chess at his level.