Efficient Performance Bounds for Primal-Dual Reinforcement Learning from Demonstrations.
Angeliki KamoutsiGoran BanjacJohn LygerosPublished in: ICML (2021)
Keyphrases
- primal dual
- reinforcement learning
- linear programming
- approximation algorithms
- algorithm for linear programming
- simplex algorithm
- convergence rate
- linear programming problems
- duality gap
- affine scaling
- interior point methods
- lower and upper bounds
- convex programming
- semidefinite programming
- variational inequalities
- learning algorithm
- convex optimization
- upper bound
- learning problems
- linear program
- convex optimization problems
- pairwise
- lower bound
- infeasible interior point