Leveraging Grammar and Reinforcement Learning for Neural Program Synthesis.
Rudy BunelMatthew J. HausknechtJacob DevlinRishabh SinghPushmeet KohliPublished in: ICLR (Poster) (2018)
Keyphrases
- program synthesis
- reinforcement learning
- fitted q iteration
- network architecture
- function approximation
- neural network
- model free
- recursive programs
- natural language
- state space
- markov decision processes
- optimal policy
- reinforcement learning algorithms
- context free grammars
- temporal difference
- multi agent
- artificial intelligence
- machine learning
- data mining
- inductive logic programming
- contextual information
- low level
- learning algorithm
- real robot
- information retrieval