An Implicit Trust Region Approach to Behavior Regularized Offline Reinforcement Learning.

Zhe Zhang Xiaoyang Tan

Published in: AAAI (2024)

Keyphrases

trust region
reinforcement learning
global optimum
column generation
least squares
optimization methods
hessian matrix
newton method
levenberg marquardt
function approximation
machine learning
genetic algorithm
state space
risk minimization
line search