Boolean Decision Rules for Reinforcement Learning Policy Summarisation.
James McCarthyRahul NairElizabeth DalyRadu MarinescuIvana DusparicPublished in: CoRR (2022)
Keyphrases
- decision rules
- reinforcement learning
- optimal policy
- policy search
- rough sets
- disjunctive normal form
- markov decision process
- action selection
- reinforcement learning problems
- rough set theory
- decision trees
- reinforcement learning algorithms
- policy gradient
- positive and negative
- actor critic
- decision table
- action space
- partially observable
- partially observable environments
- function approximation
- rule sets
- control policy
- real valued
- function approximators
- reward function
- statistical decision theory
- markov decision processes
- rule induction
- rule learning
- boolean functions
- dynamic programming
- state space
- markov decision problems
- rough sets theory
- partially observable markov decision processes
- average reward
- learning algorithm
- machine learning
- policy iteration
- temporal difference
- artificial intelligence
- real world
- real time
- fuzzy sets
- knowledge discovery
- learning problems
- decision functions
- agent learns