Policy Certificates: Towards Accountable Reinforcement Learning.
Christoph DannLihong LiWei WeiEmma BrunskillPublished in: CoRR (2018)
Keyphrases
- reinforcement learning
- optimal policy
- policy search
- markov decision process
- action selection
- state space
- markov decision processes
- reinforcement learning problems
- function approximators
- reinforcement learning algorithms
- partially observable environments
- actor critic
- policy gradient
- policy iteration
- control policies
- decision problems
- markov decision problems
- partially observable
- control policy
- action space
- function approximation
- state and action spaces
- approximate dynamic programming
- reward function
- model free
- continuous state
- learning algorithm
- approximate policy iteration
- policy evaluation
- rl algorithms
- access control
- policy gradient methods
- model free reinforcement learning
- temporal difference learning
- average reward
- optimal control
- state dependent
- finite state
- partially observable markov decision processes
- control problems
- infinite horizon
- temporal difference
- policy making
- transition model
- dynamic programming
- learning process
- multi agent
- public key