Finite-Time Convergence Rates of Decentralized Stochastic Approximation With Applications in Multi-Agent and Multi-Task Learning.
Sihan ZengThinh T. DoanJustin RombergPublished in: IEEE Trans. Autom. Control. (2023)
Keyphrases
- multi task learning
- stochastic approximation
- convergence rate
- multi agent
- policy iteration
- multi task
- reinforcement learning
- learning tasks
- gaussian processes
- step size
- monte carlo
- learning problems
- transfer learning
- high order
- theoretical analysis
- multiple agents
- single agent
- learning algorithm
- learning models
- loss function
- higher order