Hacker News new | ask | show | jobs
Visual Model-Based Reinforcement Learning as a Path Towards Generalist Robots (bair.berkeley.edu)
55 points by overlords 2750 days ago
1 comments

Those videos of the computer's prediction of events that _haven't happened yet_ are fascinating.
You might like this talk by David Cox called Predictive Coding Models of Perception: https://m.youtube.com/watch?v=P0yVuoATjzs