SCAFFLSA: Quantifying and Eliminating Heterogeneity Bias in Federated Linear Stochastic Approximation and Temporal Difference Learning.
Paul MangoldSergey SamsonovSafwan LabbiIlya LevinRéda AlamiAlexey NaumovEric MoulinesPublished in: CoRR (2024)
Keyphrases
- temporal difference learning
- stochastic approximation
- function approximation
- fixed point
- reinforcement learning
- evaluation function
- game playing
- function approximators
- policy iteration
- temporal difference
- reinforcement learning algorithms
- monte carlo
- markov decision process
- markov decision processes
- neural network
- machine learning