Work

Avoiding Side Effects in Complex Environments

Year: 2020

Type: preprint

Abstract: Reward function specification can be difficult. Rewarding the agent for making a widget may be easy, but penalizing the multitude of possible negative side effects is hard. In toy environments, Attainable Utility Preservation (AUP) avoided side effects by penalizing shifts in the ability to achieve randomly generated goals. We scale this approach to large, randomly generated environments based on Conway's Game of Life. By preserving optimal value for a single randomly generated reward function, AUP incurs modest overhead while leading the agent to complete the specified task and avoid many side effects. Videos and code are available at https://avoiding-side-effects.github.io/.

Source: arXiv (Cornell University)

DOI: 10.48550/arxiv.2006.06547

Authors: Turner, Alexander Matt, Ratzlaff, Neale, Tadepalli, Prasad

Institution: Oregon State University

Language: English

Topic: Reinforcement Learning in Robotics

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

Open Access status: green

Funders: National Science Foundation, Defense Advanced Research Projects Agency, Advanced Research Projects Agency

Awards: G1037014486, G1723748841, S&AS: INT: COLLAB: Composable and Verifiable Design for Autonomous Humanoid Robots in Space Missions, G2284091598, G8215844777, RI: Small: Speedup Learning for Online Planning Under Uncertainty

OpenAlex ID: W3035015331

Cites: 26

Cited by: 19

Related to: 10