A Definition of Open-Ended Learning Problems for Goal-Conditioned Agents.
Olivier SigaudGianluca BaldassarreCédric ColasStéphane DoncieuxRichard J. DuroNicolas Perrin-GilbertVieri Giuliano SantucciPublished in: CoRR (2023)
Keyphrases
- learning problems
- open ended
- learning tasks
- multi agent systems
- multi agent
- supervised learning
- machine learning algorithms
- autonomous agents
- learning algorithm
- machine learning
- kernel methods
- multiagent systems
- semi supervised learning
- learning outcomes
- sample complexity
- reinforcement learning
- intelligent agents
- computational learning theory
- multiple agents
- pac learning
- kernel based learning
- inquiry learning
- multi task
- data sets
- learning experience
- graph cuts
- multiple kernel learning
- object recognition
- decision trees
- real world