Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

DeepMind says reinforcement learning is ‘enough’ to reach general AI

venturebeat.com

Post reply on HN
Print view View original

201–210 of 312 posts

Re: DeepMind says reinforcement learning is ‘enough’ to reach general AI

#201

Post by treeman79 »

endtime wrote: post #192
burning_hamster wrote: post #137

Earlier quoted context omitted.

I think the title of the paper makes more sense if you consider that ten years ago, someone could have written a paper in a similar spirit with a different take on "what is enough". Back then, it would probably have been titled: "Backpropagation of errors is enough". The last ten years have shown that backpropagation -- while a crucial component -- is not enough. Personally, I would not be shocked to find out in the…

Why are forgetting and sleep relevant? If someone invented a pill that gave you a perfect memory and removed the need to sleep, would you stop being generally intelligent if you took it?

Possibly. Database look up on a million rows is very different then a lookup on a trillion. Both have solutions, but the Perl hack that is our mind may lock up on a bigger data set.

Re: DeepMind says reinforcement learning is ‘enough’ to reach general AI

#203

Post by sklearncowboy »

What proof is there that I am generally intelligent in the same context we want computers to be in order for AGI to be true?

Imagine how much farther the field would be if we stopped wasting resources on this fantasy land nonsense.

All intelligence is specialized intelligence.

Re: DeepMind says reinforcement learning is ‘enough’ to reach general AI

#204

Post by jawarner »

sklearncowboy wrote: post #203

What proof is there that I am generally intelligent in the same context we want computers to be in order for AGI to be true? Imagine how much farther the field would be if we stopped wasting resources on this fantasy land nonsense. All intelligence is specialized intelligence.

Yes, human intelligence is the benchmark for the definition of general intelligence. Certainly we can go further.

Re: DeepMind says reinforcement learning is ‘enough’ to reach general AI

#205

Post by bitL »

RL is still pretty dumb and for that extremely computationally demanding. Markov chains are like the most trivial stochastic models and I can't believe that's all it takes to get AGI. I also often wonder whether the "Deep" part of DRL is not spending most of its weights to compensate for the basic nature of Markov chains.

Re: DeepMind says reinforcement learning is ‘enough’ to reach general AI

#207

Post by woeirua »

If RL is enough, then there is no physically realizable way to actually train a RL based GAI in the near future. RL based learning requires evaluating the outcome of millions or billions of scenarios over time in order to optimize the network.

Given that requirement you'd have to either find a way to accurately model the world and all of those interactions in silicon, or you'd have to build millions of robots that can report back the results of billions of interactions each day. It's not impossible to do that, and maybe it would even be likely that we would eventually accomplish that but the cost would make it prohibitive for anyone but a nation to even attempt today. It's almost certainly outside the realm of what is possible in the near future. Maybe when robotics has progressed enough that robots are capable of interacting with the world with basic AI will we see the rise of something like a GAI.

Re: DeepMind says reinforcement learning is ‘enough’ to reach general AI

#208

Post by dkjaudyeqooe »

666lumberjack wrote: post #107
zimpenfish wrote: post #94

Earlier quoted context omitted.

> YouTube content recommendation stands out to me in particular. If that's "cutting edge ML", then going off my YouTube recommendations, we're back in another AI winter. If I watch one video from a channel I've not seen before, I'll get that channel recommended constantly even if it bears no resemblance to what I normally watch. On my Explore page, the first 22 videos (of which 8 are Fortnite-related!) hold no intere…

How often do you use YouTube? Personally I am a very heavy user and in my experience the obsession with a new video kind you watch only lasts for a few recommendations unless you lean into it. I would guess about two thirds of the channels I consistently watch I originally discovered through algorithm recommendations. I think it works extremely well.

That's because you fit into YT's conception of how viewers behave. For people who don't fit into "normal"-ish behaviour it has little utility.

For me, probably 90% of what I watch I'm not interested in and often I'm repelled by. This is because I mostly watch to find out what things I'm not familiar with are.

For example let's say I'm a liberal. I'm not going to watch liberal political videos because I know generally what they're going to say and I don't need my political views stroked in order to be happy. But I will watch various other political videos, no matter how extreme or not, so I can be at least a little familiar with their behaviour and views.

YT can't cope with this. To their systems I seem to be randomly picking videos with no correlation with the subject matter or other users and no reinforcing pattern. It just gives up and recommends things based on the behaviour of the general population, as if they had no data on me at all.

Re: DeepMind says reinforcement learning is ‘enough’ to reach general AI

#210

Post by rubatuga »

webmaven wrote: post #40
jjjdjjddddsfsd wrote: post #25

Are the just reformulating the principles of evolution in digital terms, and essentially not providing any new insights at all? Yes, intelligence has been created by evolution. That doesn't imply that any system that is subject to evolutionary forces will lead to the creation of intelligence (and not within a reasonable timeframe, either). The challenge is to create a system that is capable of evolving intelligence.…

> Afaik some biologists even think that the evolution of intelligence was rather unlikely and would not necessarily happen again under the same circumstances as on earth. Hmm. Can you provide a pointer to those biologists? AFAIK, high intelligence has arisen more than once on Earth (Hominoids, Cetaceans, Octopuses), so I'm somewhat skeptical of that claim, but perhaps they're construing intelligence more narrowly (ie…

Well that has a prior on life even existing in the first place
Post reply on HN