The world's leading venue for collaborative research in theoretical computer science. Follow us at http://YouTube.com/SimonsInstitute.
Simons Institute for the Theory of Computing
Loading...
1/2 The Return of RNNs. "AI is still lacking a good memory mechanism...We are in this era of 'attention is all you need' but I don't think that's going to cut it," said @phillipisola.bsky.social of @mit.edu at the Simons Institute workshop on Topics in Intelligence: World Models and Social Reasoning
1/4 World models are all the rage in both neuroscience and AI. But does the brain really build world models? Jack Gallant (@ucberkeleyofficial.bsky.social) asked at the Simons Institute workshop on Topics in Intelligence: World Models and Social Reasoning. Video: simons.berkeley.edu/talks/jack-g...
2/2 But "Bayes theory is just a law of the universe. Any system that works optimally in the world has to obey Bayes theorem. So, it’s going to be really hard to find a situation where Bayes theorem isn’t a good model," said @gallantlab.org of @ucberkeleyofficial.bsky.social at the Simons Institute.
2/2 "We already have the solution..Everybody knew that recurrent neural networks [would] eventually have to come back...They solve the memory problem from the ground up," said @phillipisola.bsky.social of @mit.edu at the Simons Institute. Video: simons.berkeley.edu/talks/philli...
1/2 Generative world model (GWM) theories are correlated with theories that aren't world model theories, such as the Bayesian brain theory, said @gallantlab.org at the Simons Institute workshop on Topics in Intelligence: World Models and Social Reasoning. simons.berkeley.edu/talks/jack-g...
2/4 "Is the brain just picking up on correlations and predicting the next state of the world based on the correlation structure of the world?"—Jack Gallant of @ucberkeleyofficial.bsky.social at the Simons Institute workshop on Topics in Intelligence: World Models and Social Reasoning
3/4 "Or is the brain instantiating some underlying model of the causes of the events that occur in the world that's low-dimensional, highly compressed, highly efficient representation of the structure of the world," asked Jack Gallant of @ucberkeleyofficial.bsky.social at the Simons Institute.
4/4 "Nobody knows and the evidence sucks. Nobody has any idea in the brain which of these two groups of theories is true, and it's really a horror show of sad, vague results," @gallantlab.org at the Simons Institute workshop on Topics in Intelligence: World Models and Social Reasoning
