The real obstacles to the path to AI is that we don't even have the right questions to ask and people are looking for answers to the wrong questions and announcing the next big thing in AI.
Obstacles on the Path to AI
21–30 of 31 posts
Re: Obstacles on the Path to AI
#22Earlier quoted context omitted.
Yes, I should have added the context about the scalar reward, but I'm not sure why that changes anything. This may seem like a naive question, but it's sincere: What makes a scalar reward less effective at modifying a Q function than a scalar error that's used in backprop and assigned to a neural network's coefficients?
supervised learning: (after millions of operations you had just performed) out of 1000 you predicted label #136, but actually the true result for this input was label #25. reinforcement learning: (after millions of operations you had just performed) out of 1000 you predicted label #136. That's not right, but I won't tell you what you should have done. Also, it could have been right, but maybe you had screwed up somet…
This is an overly restrictive view of RL. Yann's claim about the potential utility of RL, taken at face value, is clearly false. Backprop on a deterministic computation graph is equivalent to deterministic policy gradient in the same graph, where the reward is given by the value of the optimization objective.
Re: Obstacles on the Path to AI
#23"There is no way in hell that you can learn billions of parameters with RL." I really love LeCun's provocative stances, but I get suspicious when people talk about impossibilities. RL is making huge strides. People adopted the same tone with neural nets years ago, and LeCun proved them wrong...
In biology, it's a combination of Reinforcement learning and Unsupervised learning. But he probably means just RL on its own. Although I'm not sure if that statement would be correct.
Re: Obstacles on the Path to AI
#24Re: Obstacles on the Path to AI
#25Re: Obstacles on the Path to AI
#26Everytime I read something like this, I get sad that I don't understand most of it. But then I get happy because at least I understand a little bit :)
Reading some random slides from a specialist talk is not exactly the easiest way to learn this stuff.
Most often the slides alone are useless, they are only supposed to be a support for the talk. Talks recordings (or transcripts) are more useful, you can only get a high level idea of what the talk is about from slides.
I would be awesome to have a platform where you get recommendations of what to learn / read or online courses in order to understand a given talk.
Re: Obstacles on the Path to AI
#27Although I'm working on deep neural nets, this material is too advanced to me. Looks like deep nets + bayesian reasoning is the next big thing.
Pick up this book. It's fantastic. https://mitpress.mit.edu/books/probabilistic-graphical-model...
The title sounds familiar, it's also a course on coursera:
https://www.coursera.org/course/pgm
Last session was in 2013 though.
Re: Obstacles on the Path to AI
#28Earlier quoted context omitted.
Reading some random slides from a specialist talk is not exactly the easiest way to learn this stuff.
Most often the slides alone are useless, they are only supposed to be a support for the talk. Talks recordings (or transcripts) are more useful, you can only get a high level idea of what the talk is about from slides. I would be awesome to have a platform where you get recommendations of what to learn / read or online courses in order to understand a given talk.
Yes absolutely! A platform where not only the answers are praised, but also curious questions.
Re: Obstacles on the Path to AI
#29That's only the slides. Is there a video of the talk?(assuming there is a talk, that is)
I'd love to see the video too... LeCun is probably most interesting person to hear in this topic...
In case you didn't see the comment originally (I didn't either), someone else posted a link.
http://techtalks.tv/talks/whats-wrong-with-deep-learning/616...
Re: Obstacles on the Path to AI
#30The talk: http://techtalks.tv/talks/whats-wrong-with-deep-learning/616...
This is not the same talk. The "what's wrong with deep learning" talk was given in CVPR 2015. The slides linked in this HN post was presented in BayLearn: Bay Area Machine Learning Symposium.