Neuro-Symbolic Reinforcement Learning with First-Order Logic.
Daiki KimuraMasaki OnoSubhajit ChaudhuryRyosuke KohitaAkifumi WachiDon Joven AgravanteMichiaki TatsuboriAsim MunawarAlexander GrayPublished in: EMNLP (1) (2021)
Keyphrases
- first order logic
- reinforcement learning
- expressive power
- propositional logic
- theorem proving
- theorem prover
- inference rules
- knowledge representation
- probabilistic reasoning
- state space
- inductive logic programming
- representation language
- transitive closure
- relational algebra
- horn clauses
- markov networks
- predicate calculus
- probabilistic graphical models
- optimal policy
- learning algorithm
- markov decision processes
- expert systems
- anti unification
- proof procedure
- markov logic
- quantifier free
- binary decision diagrams
- constraint databases
- transfer learning
- graphical models