Source Themes

We present a unified framework for intrinsic motivation and reward shaping: they signal the value of the RL agent’s state, which combines external state and past experience. Rewards based on potentials over the learning agent’s state provably avoid reward hacking.

We demonstrate how to generalize over a combinatorially large space of rearrangement tasks from only pixel observations by constructing from video demonstrations a factorized transition graph over entity state transitions that we use for control.

We demonstrate how to generalize over a combinatorially large space of rearrangement tasks from only pixel observations by constructing from video demonstrations a factorized transition graph over entity state transitions that we use for control.

Quick, draw a sandwich! What does it look like? If you were from South Korea, you may have added pickled cucumber, or if you were from …