- Title
- Potential-based multiobjective reinforcement learning approaches to low-impact agents for AI safety
- Creator
- Vamplew, Peter; Foale, Cameron; Dazeley, Richard; Bignold, Adam
- Date
- 2021
- Type
- Text; Journal article
- Identifier
- http://researchonline.federation.edu.au/vital/access/HandleResolver/1959.17/176481
- Identifier
- vital:15049
- Identifier
-
https://doi.org/10.1016/j.engappai.2021.104186
- Identifier
- ISBN:0952-1976 (ISSN)
- Abstract
- The concept of impact-minimisation has previously been proposed as an approach to addressing the safety concerns that can arise from utility-maximising agents. An impact-minimising agent takes into account the potential impact of its actions on the state of the environment when selecting actions, so as to avoid unacceptable side-effects. This paper proposes and empirically evaluates an implementation of impact-minimisation within the framework of multiobjective reinforcement learning. The key contributions are a novel potential-based approach to specifying a measure of impact, and an examination of a variety of non-linear action-selection operators so as to achieve an acceptable trade-off between achieving the agent's primary task and minimising environmental impact. These experiments also highlight a previously unreported issue with noisy estimates for multiobjective agents using non-linear action-selection, which has broader implications for the application of multiobjective reinforcement learning. © 2021
- Publisher
- Elsevier Ltd
- Relation
- Engineering Applications of Artificial Intelligence Vol. 100, no. (2021), p.
- Rights
- All metadata describing materials held in, or linked to, the repository is freely available under a CC0 licence
- Rights
- Copyright © 2021 Elsevier Ltd. All rights reserved.
- Rights
- Open Access
- Subject
- 08 Information and Computing Sciences; 09 Engineering; AI safety; Low-impact agents; Multiobjective reinforcement learning; Potential-based rewards; Reward engineering; Safe reinforcement learning; Side-effects
- Full Text
- Reviewed
- Hits: 2862
- Visitors: 2467
- Downloads: 137
| Thumbnail | File | Description | Size | Format | |||
|---|---|---|---|---|---|---|---|
| View Details Download | SOURCE1 | Accepted version | 1 MB | Adobe Acrobat PDF | View Details Download |