BF++: a language for general-purpose program synthesis

Abstract

Most state of the art decision systems based on Reinforcement Learning (RL) are data-driven black-box neural models, where it is often difficult to incorporate expert knowledge into the models or let experts review and validate the learned decision mechanisms. Knowledge-insertion and model review are important requirements in many applications involving human health and safety. One way to bridge the gap between data and knowledge driven systems is program synthesis: replacing a neural network that outputs decisions with a symbolic program generated by a neural network or by means of genetic programming. We propose a new programming language, BF++, designed specifically for automatic programming of agents in a Partially Observable Markov Decision Process (POMDP) setting and apply neural program synthesis to solve standard OpenAI Gym benchmarks.

8+2 pages (paper+references)

References in corpus (11)

• A Brief Survey of Deep Reinforcement Learning
• SQLNet: Generating Structured Queries From Natural Language Without Reinforcement Learning
• Neural-Symbolic Learning and Reasoning: A Survey and Interpretation
• Reinforcement Learning in Healthcare: A Survey
• TerpreT: A Probabilistic Programming Language for Program Induction
• Neural-Guided Deductive Search for Real-Time Program Synthesis from Examples
• Reinforcement Learning Applications
• Synthetic Datasets for Neural Program Synthesis
• Recent Advances in Neural Program Synthesis
• DAQN: Deep Auto-encoder and Q-Network
• Zooming for Efficient Model-Free Reinforcement Learning in Metric Spaces