Multi-fitness learning for behavior-driven cooperation
| Autor: | Kagan Tumer, Reid Christopher, Connor Yates |
|---|---|
| Rok vydání: | 2020 |
| Předmět: |
Sequence
business.industry Computer science SIGNAL (programming language) Evolutionary learning 0102 computer and information sciences 02 engineering and technology 01 natural sciences Task (project management) 010201 computation theory & mathematics Order (exchange) 0202 electrical engineering electronic engineering information engineering 020201 artificial intelligence & image processing Artificial intelligence business |
| Zdroj: | GECCO |
| DOI: | 10.1145/3377930.3390220 |
| Popis: | Evolutionary learning algorithms have been successfully applied to multiagent problems where the desired system behavior can be captured by a single fitness signal. However, the complexity of many real world applications cannot be reduced to a single number, particularly when the fitness (i) arrives after a lengthy sequence of actions, and (ii) depends on the joint-action of multiple teammates. In this paper, we introduce the multi-fitness learning paradigm to enable multiagent teams to identify which fitness matters when in domains that require long-term, complex coordination. We demonstrate that multi-fitness learning efficiently solves a cooperative exploration task where teams of rovers must coordinate to observe various points of interest in a specific but unknown order. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|