This repository contains the code and results from the TAILOR 2020 paper "Synthesising Reinforcement Learning Policies
through Set-Valued Inductive Rule Learning" by Coppens, Steckelmacher, Jonker and Nowé.
Through Set-Valued Inductive Rule Learning" by Coppens, Steckelmacher, Jonker and Nowé.
The final authenticated publication is available online at [https://doi.org/10.1007/978-3-030-73959-1_15](https://doi.org/10.1007/978-3-030-73959-1_15)
We developed a strategy to translate black-box neural network policies from the Deep RL algorithms into a
list of inductive rules, while taking sets of "equally good" actions into account.
