Timezone: »
Despite the remarkable success of Deep RL in learning control policies from raw pixels, the resulting models do not generalize. We demonstrate that a trained agent fails completely when facing small visual changes, and that fine-tuning---the common transfer learning paradigm---fails to adapt to these changes, to the extent that it is faster to re-train the model from scratch. We show that by separating the visual transfer task from the control policy we achieve substantially better sample efficiency and transfer behavior, allowing an agent trained on the source task to transfer well to the target tasks. The visual mapping from the target to the source domain is performed using unaligned GANs, resulting in a control policy that can be further improved using imitation learning from imperfect demonstrations. We demonstrate the approach on synthetic visual variants of breakout game, as well as on transfer between subsequent levels of Road Fighter, a Nintendo car-driving game. A visualization of our approach can be seen in \url{https://streamable.com/ymv8v} and \url{https://streamable.com/fvy1d}.
Author Information
Shani Gamrian (Bar Ilan University)
Yoav Goldberg
Related Events (a corresponding poster, oral, or spotlight)
-
2019 Poster: Transfer Learning for Related Reinforcement Learning Tasks via Image-to-Image Translation »
Fri. Jun 14th 01:30 -- 04:00 AM Room Pacific Ballroom
More from the Same Authors
-
2022 Poster: Linear Adversarial Concept Erasure »
· Michael Twiton · Yoav Goldberg · -
2022 Spotlight: Linear Adversarial Concept Erasure »
· Michael Twiton · Yoav Goldberg ·