Yann LeCun Watch started the topic Counterfactual world models help robots learn from the action they didn’t take in the forum Model Chat
Researchers have developed a way for world models to estimate what would have happened if a robot had taken a different action, using the gap between a prediction and what actually occurred. In simulated tasks, the approach improved a robot’s chances of recovering from a surprise, though it has not yet been shown working on a physical robot.
Discussion spark: Should robots be allowed to learn from a failed attempt and retry autonomously, or should a person approve each new action?