Reincarnating RL: Reusing Prior Computation to Accelerate Progress
Nov 28, 2022
Speakers
Rishabh Agarwal
Speaker · 2 followers
Max Schwarzer
Speaker · 1 follower
Pablo Samuel Castro
Speaker · 1 follower
Aaron Courville
Speaker · 3 followers
Marc Bellemare
Speaker · 3 followers
About
Learning tabula rasa, that is without any prior knowledge, is the prevalent workflow in reinforcement learning (RL) research. However, RL systems, when applied to large-scale settings, rarely operate tabula rasa. Such large-scale systems undergo multiple design or algorithmic changes during their development cycle and use ad hoc approaches for incorporating these changes without re-training from scratch, which would have been prohibitively expensive. Additionally, the inefficiency of deep RL typically excludes researchers without access to industrial-scale resources from tackling computationally-demanding problems. To address these issues, we present reincarnating RL as an alternative workflow or class of problem settings, where prior computational work (e.g., learned policies) is reused or transferred between design iterations of an RL agent, or from one RL agent to another. As a step towards enabling reincarnating RL from any agent to any other agent, we focus on the specific setting of efficiently transferring an existing sub-optimal policy to a standalone value-based RL agent. We find that existing approaches fail in this setting and propose a simple algorithm to address their limitations. Equipped with this algorithm, we demonstrate reincarnating RL's gains over tabula rasa RL on Atari 2600 games, a challenging locomotion task, and the real-world problem of navigating stratospheric balloons. Overall, this work argues for an alternate approach to RL research, which we believe could significantly improve real-world RL adoption and help democratize it further.Learning tabula rasa, that is without any prior knowledge, is the prevalent workflow in reinforcement learning (RL) research. However, RL systems, when applied to large-scale settings, rarely operate tabula rasa. Such large-scale systems undergo multiple design or algorithmic changes during their development cycle and use ad hoc approaches for incorporating these changes without re-training from scratch, which would have been prohibitively expensive. Additionally, the inefficiency of deep RL typ…
Organizer
NeurIPS 2022
Account · 954 followers
Like the format? Trust SlidesLive to capture your next event!
Professional recording and live streaming, delivered globally.
Sharing
Recommended Videos
Presentations on similar topic, category or speaker
Alignment-guided Temporal Attention for Video Action Recognition
Watch later
Yizhou Zhao, …
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
BayesPCN: A Continually Learnable Predictive Coding Associative Memory
Watch later
Jinsoo Yoo, …
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
Temporal Graph Learning: Some Challenges and Recent Directions
Watch later
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
Towards Reasoning-Aware Explainable VQA
Watch later
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
Is Model Ensemble Necessary? Model-based RL via a Single Model with Lipschitz Regularized Value Function
Watch later
Ruijie Zheng, …
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
Learning to Scaffold: Optimizing Model Explanations for Teaching
Watch later
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%