White-Box Adversarial Policies in Deep Reinforcement Learning.
Stephen CasperDylan Hadfield-MenellGabriel KreimanPublished in: CoRR (2022)
Keyphrases
- white box
- reinforcement learning
- black box
- optimal policy
- policy search
- markov decision process
- source code
- multi agent
- control policies
- reward function
- fitted q iteration
- partially observable markov decision processes
- markov decision processes
- markov decision problems
- reinforcement learning agents
- reinforcement learning algorithms
- function approximation
- policy gradient methods
- state space
- test data
- hierarchical reinforcement learning
- source code metrics
- model free
- temporal difference
- test cases
- continuous state
- control policy
- decision problems
- partially observable
- neural network
- learning algorithm
- function approximators
- multi agent reinforcement learning
- decentralized control
- multiagent reinforcement learning
- finite state
- transfer learning
- action space
- feature selection
- policy iteration
- infinite horizon
- databases
- database