BF++: a language for general-purpose program synthesis
This addresses the need for more interpretable and knowledge-integratable decision systems in applications involving human health and safety.
The paper tackles the problem of incorporating expert knowledge and enabling model review in reinforcement learning systems by proposing BF++, a new programming language for program synthesis in partially observable Markov decision processes, and applies neural program synthesis to solve standard OpenAI Gym benchmarks.
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.