Uncertainty Estimation Using Riemannian Model Dynamics for Offline Reinforcement Learning.

Guy Tennenholtz Shie Mannor

Published in: NeurIPS (2022)

Keyphrases

reinforcement learning
computational model
probability distribution
objective function
parameter estimation
mathematical model
experimental data
decision theoretic
real time
parametric models
function approximation
uncertain data
closed form
kalman filter
dynamical systems
state space
learning algorithm