- Title
- On the limitations of scalarisation for multi-objective reinforcement learning of Pareto fronts
- Creator
- Vamplew, Peter; Yearwood, John; Dazeley, Richard; Berry, Adam
- Date
- 2008
- Type
- Text; Conference paper
- Identifier
- http://researchonline.federation.edu.au/vital/access/HandleResolver/1959.17/61003
- Identifier
- vital:3742
- Identifier
-
https://doi.org/10.1007/978-3-540-89378-3_37
- Identifier
- ISBN:9783540893776
- Abstract
- Multiobjective reinforcement learning (MORL) extends RL to problems with multiple conflicting objectives. This paper argues for designing MORL systems to produce a set of solutions approximating the Pareto front, and shows that the common MORL technique of scalarisation has fundamental limitations when used to find Pareto-optimal policies. The work is supported by the presentation of three new MORL benchmarks with known Pareto fronts.
- Publisher
- Auckland, New Zealand : Springer
- Relation
- Paper presented at 21st Australasian Joint Conference on Artificial Intelligence, Auckland, New Zealand : 1st-5th December 2008 Vol. 5360, p. 372-378
- Rights
- Copyright Springer
- Rights
- This metadata is freely available under a CCO license
- Subject
- 0801 Artificial Intelligence and Image Processing; Multiobjective; Reinforcement learning; Scalarisation; Pareto fronts
- Hits: 2075
- Visitors: 1992
- Downloads: 1
| Thumbnail | File | Description | Size | Format |
|---|