Convergence Rates for Stochastic Approximation: Biased Noise with Unbounded Variance, and Applications.

Rajeeva L. Karandikar M. Vidyasagar

Published in: CoRR (2023)

Keyphrases

stochastic approximation
convergence rate
policy iteration
convergence speed
learning rate
monte carlo
primal dual
step size
numerical stability
upper bound
reinforcement learning
optical flow
theoretical guarantees
temporal difference learning
learning algorithm
gaussian kernels
neural network