Upper Confidence Primal-Dual Reinforcement Learning for CMDP with Adversarial Loss.
Shuang QiuXiaohan WeiZhuoran YangJieping YeZhaoran WangPublished in: NeurIPS (2020)
Keyphrases
- primal dual
- reinforcement learning
- reactive planning
- linear programming
- affine scaling
- convex optimization
- convergence rate
- linear program
- interior point methods
- linear programming problems
- multi agent
- approximation algorithms
- variational inequalities
- algorithm for linear programming
- simplex algorithm
- interior point algorithm
- semidefinite programming
- state space
- interior point
- markov decision processes
- dynamic programming
- duality gap
- dual formulation
- line search
- model free
- column generation
- simplex method
- control structure
- machine learning
- optimal policy
- super resolution
- multiresolution
- optimal solution
- image processing
- knowledge base
- learning algorithm