Learning to Repair Infeasible$^*$ Problems with Deep Reinforcement Learning on Graphs

Mehdi Zouitine, Ahmad Berjaoui, Agnès Lagnoux, Pellegrini, Emmanuel Rachelson

Confirmation: I have read and agree with the workshop's policy on behalf of myself and my co-authors.

Tracks: Main Track

Keywords: Infeasibility Analysis, Graph Neural Networks, Learning Based Heuristics, Linear Feasibility, Boolean Satisfiability, Deep Reinforcement Learning

TL;DR: Learning to Repair Infeasible Problems with Deep Reinforcement Learning on Graphs

Abstract: In the last few years, deep learning has demonstrated significant potential in Operations Research across various tasks. In this work, we tackle the problem of repairing infeasible constraint satisfaction problems by introducing a novel deep reinforcement learning approach. Our method leverages graph deep learning to represent infeasible problems, utilizing a graph representation of Constraint Satisfaction Problems. By employing bipartite graph neural networks to encode the constraints of these problems, we train a deep learning agent to identify and extract a subset of constraints that restores feasibility solely from the reward signal, requiring no labeled data. We evaluate our approach using several bipartite graph neural network architectures and demonstrate its effectiveness in two domains: maximizing feasibility in Linear Programs and maximizing satisfiability in Boolean satisfiability problems. Our results show that the agent is competitive with existing heuristics in both solution quality and computational efficiency across these domains. An open source implementation of our methods is available at https://github.com/MehdiZouitine/Learning_to_repair_infeasible_problems_with_DRL_and_GNN.

Submission Number: 48