Yoshua Bengio: Pausing More Powerful AI Models and His Work on World Models
Eye On A.I.
00:00
Bayesian Structure Learning With a Geranted Flow Networks
A theory is like a likelihood function. So it's something or an energy function in terms of young bikes. A theory gives us a checkable quantity about how well it fits the data. With standard reinforcement learning, it will pick one theory and be happy. You want to have all the theories somehow. And if a theory fits the data twice as good, maybe you want to sample it twice more often. That's the vision we are thinking about how you should be rational in your decision-making.
Transcript
Play full episode
