Ecole: A Gym-like Library for Machine Learning in Combinatorial Optimization Solvers

Antoine Prouvost, Justin Dumouchelle, Lara Scavuzzo, Maxime Gasse, Didier Chételat, Andrea Lodi

Keywords: machine learning, combinatorial optimization, integer programming, control, open-source, MDP, gym, reinforcement learning

TL;DR: Ecole is a new library that exposes control problems arising in general-purpose combinatorial optimization solvers as MDP environments in Python.

Abstract: We present Ecole, a new library to simplify machine learning research for combinatorial optimization. Ecole exposes several key decision tasks arising in general-purpose combinatorial optimization solvers as control problems over Markov decision processes. Its interface mimics the popular OpenAI Gym library and is both extensible and intuitive to use. We aim at making this library a standardized platform that will lower the bar of entry and accelerate innovation in this growing field. Documentation and code can be found at https://www.ecole.ai.

Community Implementations: [ 1 code implementation](https://www.catalyzex.com/paper/arxiv:2011.06069/code)