Yann LeCun's comment on AlphaGo and true AI • Hacker News Board

Re: Yann LeCun's comment on AlphaGo and true AI

#181

Post by javajosh » Mon, Mar 14, 2016, 5:41 PM UTC

wpietri wrote: post #152

javajosh wrote: post #79

Earlier quoted context omitted.

Cool, thanks. I think the idea is also important to Zoltan Torey, who believed that "consciousness" is essentially a simulation of a kind of arm in our mind. ("The Conscious Mind"). But I've never heard of anyone attempting to train an AI within a simulated world. The nice thing about doing that is that aren't limited by physics of robotics. I suspect that for basic intelligence stuff, relatively lo-fidelity simulati…

> But I've never heard of anyone attempting to train an AI within a simulated world. Ah, then you might want to follow the trail from SHRLDU: http://hci.stanford.edu/winograd/shrdlu/ https://en.wikipedia.org/wiki/SHRDLU

Thanks for the link! That's pretty close to what I mean; but I was thinking more like an embodied AI living in a simulated world. So, for example, you might start from a Counter-Strike bot, input to the bot would be it's 'camera', and output would be it's position and actions.

Re: Yann LeCun's comment on AlphaGo and true AI

#182

Post by ewjordan » Mon, Mar 14, 2016, 5:42 PM UTC

argonaut wrote: post #10

jacquesm wrote: post #6

Earlier quoted context omitted.

> I know this will be controversial, psychology Why would that be controversial? It seems to make extremely good sense and even though it may be doubted in some circles I think that most people involved in AI research are painfully aware of our limited understanding of our own psyche.

I disagree with hacknat (and the reason I suspect it's controversial) because psychology is generally far too high-level (e.g. focused on things like traits and high-level behavior and vague "perceptions") and has far too little empirical rigor for engineers to be able to build any sort of actual system from insights gleaned from psychology. The recent replication crisis in psychology does little to help this reputat…

Even now, deep learning is basically 1980s-era backdrop with a few tweaks and better hardware - it turns out that despite decades of backlash against it, plain old backdrop pretty much is the shit, as long as you have a lot of compute power and a few tricks up your sleeve!

Re: Yann LeCun's comment on AlphaGo and true AI

#183

Post by daveguy » Mon, Mar 14, 2016, 5:44 PM UTC

pavanky wrote: post #36

Can someone more knowledgeable explain why biological systems are considered unsupervised instead of reinforcement based systems? While it seems intuitive that most individual "intelligent" systems in animals can be seen as unsupervised, isn't life itself driven in a reinforced manner?

In AI the definition of reinforcement learning vs unsupervised is that reinforcement requires external identification of the correct result to compare with the given result.

If going through life you had artificial reality display that flashed "apple" in front of every apple and "orange" in front of every orange you would have reinforcement learning. There is certainly a major component of reinforcement in teaching, but that does not account for all of learning.

With humans we are able to transfer concepts to different situations without examples and there are a lot of categories that are identified without any explicitly reinforced examples.

Re: Yann LeCun's comment on AlphaGo and true AI

#184

Post by daveguy » Mon, Mar 14, 2016, 5:44 PM UTC

pavanky wrote: post #36

Can someone more knowledgeable explain why biological systems are considered unsupervised instead of reinforcement based systems? While it seems intuitive that most individual "intelligent" systems in animals can be seen as unsupervised, isn't life itself driven in a reinforced manner?

In AI the definition of reinforcement learning vs unsupervised is that reinforcement requires external identification of the correct result to compare with the given result. If going through life you had artificial reality display that flashed "apple" in front of every apple and "orange" in front of every orange you would have reinforcement learning. There is certainly a major component of reinforcement in teaching, but that does not account for all of learning. There is a lot of transfer of concepts to different situations and there are a lot of categories that are identified without any explicitly reinforced examples.

Re: Yann LeCun's comment on AlphaGo and true AI

#185

Post by nabla9 » Mon, Mar 14, 2016, 5:48 PM UTC

sabertoothed wrote: post #155

nabla9 wrote: post #136

Earlier quoted context omitted.

There is no enough reinforcement volume to learn anything complex. If child gets external reward for every waking moment until she is 12 years old, it's just 4.2 million signals. Reinforcement learning works for fine motor control and other tasks where the feedback loop is tight and immediate. Reinforcement and conditioning can also modulate high level cognition and behavior, but it's not the secret sauce of learning…

I could not follow your argument. Could you elaborate? (Honest question)

Reinforcement learning is learning by interacting with an environment. RL agent learns from the consequences of its actions.

Biological system don't live long enough to get enough feedback to learn complex behavior trough consequences. Animal or human must be able to generalize and categorize what they have learned correctly without external feedback teaching it how to derive the function that's doing it.

For example, if you want to learn how to tie a complex knot and learn it trough trial and error you might have try it million times if you improve your behavior mainly trough consequences of your actions. In practice you probably try only 5-10 times before you learn to do it and it involves pausing and looking at the problem. There is some kind of unsupervised model building happening that is not involving external input.

Re: Yann LeCun's comment on AlphaGo and true AI

#186

Post by kafkaesq » Mon, Mar 14, 2016, 5:50 PM UTC

A beautifully concise statement of an incredibly common misconception as to the current state of the field.

Re: Yann LeCun's comment on AlphaGo and true AI

#187

Post by Houshalter » Mon, Mar 14, 2016, 5:55 PM UTC

sampo wrote: post #72

Houshalter wrote: post #61

No one is claiming that alphaGo is close to AGI. At least not anyone that understands the methods it uses. What alphaGo is, is an example of AI progress. There has been a rapid increase in progress in the field of AI. We are still a ways away from AGI, but it's now in sight. Just outside the edge of our vision. Almost surely within our lifetime, at this rate.

> We are still a ways away from AGI, but it's now in sight. These people in 1956 also thought that AGI was in sight: https://en.wikipedia.org/wiki/Dartmouth_Conferences

Not most of them, and not with any certainty. The Dartmouth conference was just the very first investigation into whether it was even possible and what it would take. Of course they didn't know when AGI would be developed. The computers they had access to were smaller than my calculator!

Now we have massive computers and 60 years of progress. And lots of real world successes of AI technology. Surely we are in a much better position to predict than people of 60 years ago!

Re: Yann LeCun's comment on AlphaGo and true AI

#188

Post by maxander » Mon, Mar 14, 2016, 5:55 PM UTC

It sounds reversed to me- shouldn't the "cherry" be supervised learning and the "icing" be reinforcement learning? At least insofar as reinforcement learning is closer to the "cake" of unsupervised learning, as there is less feedback required for a reinforcement learning system to work (a binary correctness signal rather than an n-dimensional label signal.)

It might also be argued that most "unsupervised learning" in animals can be broken down into a relatively simple unsupervised segment (e.g., an "am I eating nice food" partition function) and a more complicated reinforcement segment (e.g. a "what is the best next thing to do to obtain nice food?" function.) I'm sure someone like Yann LeCun is familiar with such arguments, though.

Re: Yann LeCun's comment on AlphaGo and true AI

#189

Post by daveguy » Mon, Mar 14, 2016, 5:57 PM UTC

justicezyx wrote: post #53

rm999 wrote: post #45

Earlier quoted context omitted.

The brain needs to form structure before any sort of reinforcement learning is even possible. This structure comes from an "unsupervised" process we don't understand well.

Isn't it from millions year of natural selection? The reason why a human brain can think, is just by chance.

Personally I think a lot of it has to do with evolution. We certainly have different sections of the brains that are more utilized in different tasks, auditory, visual, motor, etc. I think it is likely a lot of the initial feedback is preset -- the instinctual positive and negative responses of certain stimulus.

Re: Yann LeCun's comment on AlphaGo and true AI

#190

Post by deegles » Mon, Mar 14, 2016, 5:58 PM UTC

I would like future competitions between AIs and humans to have a "power budget" for training and during gameplay. For example, a chess grandmaster that has played for 20 years would have spent X amount of energy training. The AI should get an equivalent budget to train with. During gameplay, the AI would get the same 20 watts [1] that a human has. This would drive the development of more efficient hardware instead of throwing power at the problem :)

[1] http://www.popsci.com/technology/article/2009-11/neuron-comp...