Towards Interpretable Policies in Multi-agent Reinforcement Learning Tasks.
Marco CrespiLeonardo Lucio CustodeGiovanni IaccaPublished in: BIOMA (2022)
Keyphrases
- learning tasks
- multi agent
- reinforcement learning
- transfer learning
- optimal policy
- machine learning
- learning problems
- learning algorithm
- supervised learning
- meta learning
- multi task
- multi label
- learning experience
- multi task learning
- machine learning algorithms
- function approximation
- state space
- reinforcement learning algorithms
- multiple agents
- single agent
- reward function
- hypothesis space
- data sets
- markov decision processes
- metric learning
- similarity measure
- kernel based learning methods
- kernel methods
- natural language
- learning process
- kernel learning
- labeled and unlabeled data
- control policy
- multitask learning