Instance-wise algorithm configuration with graph neural networks
This work addresses the challenge of efficiently solving MILPs for combinatorial optimization practitioners, representing an incremental advance by applying a known method to a specific competition task.
The paper tackled the problem of configuring the SCIP solver for mixed integer linear programs (MILPs) by training a graph neural network to predict instance-specific configurations, resulting in performance improvements of 12%, 35%, and 8% over the default solver on three benchmarks.
We present our submission for the configuration task of the Machine Learning for Combinatorial Optimization (ML4CO) NeurIPS 2021 competition. The configuration task is to predict a good configuration of the open-source solver SCIP to solve a mixed integer linear program (MILP) efficiently. We pose this task as a supervised learning problem: First, we compile a large dataset of the solver performance for various configurations and all provided MILP instances. Second, we use this data to train a graph neural network that learns to predict a good configuration for a specific instance. The submission was tested on the three problem benchmarks of the competition and improved solver performance over the default by 12% and 35% and 8% across the hidden test instances. We ranked 3rd out of 15 on the global leaderboard and won the student leaderboard. We make our code publicly available at \url{https://github.com/RomeoV/ml4co-competition} .