Leveraging Grammar and Reinforcement Learning for Neural Program Synthesis.

Rudy Bunel Matthew J. Hausknecht Jacob Devlin Rishabh Singh Pushmeet Kohli

Published in: CoRR (2018)

Keyphrases

program synthesis
reinforcement learning
fitted q iteration
network architecture
function approximation
neural network

natural language
inductive logic programming
recursive programs
sensory inputs
reinforcement learning algorithms
model free

multi agent
state space
transfer learning
context free grammars
machine learning

associative memory
markov decision processes
information extraction
mobile robot
feature space