Performative Reinforcement Learning in Gradually Shifting Environments.
Ben RankStelios TriantafyllouDebmalya MandalGoran RadanovicPublished in: CoRR (2024)
Keyphrases
- reinforcement learning
- real world
- dynamic environments
- markov decision processes
- function approximation
- machine learning
- control problems
- state space
- highly dynamic
- multi agent environments
- search engine
- reinforcement learning algorithms
- direct policy search
- database
- action space
- partially observable
- temporal difference
- optimal policy
- evolutionary algorithm
- bayesian networks
- neural network
- databases
- data sets