Warmstarting of Model-based Algorithm Configuration

TL;DR

This paper proposes a warmstarting approach for algorithm configuration that leverages previous benchmark data to significantly accelerate and improve the tuning process of solvers on new problem instances.

Contribution

It introduces two methods to warmstart model-based AC using prior performance data, enhancing efficiency and solution quality.

Findings

Achieved up to 165-fold speedups in configuration time.

Found better configurations with the same compute budget.

Demonstrated effectiveness on multiple SAT instance sets.

Abstract

The performance of many hard combinatorial problem solvers depends strongly on their parameter settings, and since manual parameter tuning is both tedious and suboptimal the AI community has recently developed several algorithm configuration (AC) methods to automatically address this problem. While all existing AC methods start the configuration process of an algorithm A from scratch for each new type of benchmark instances, here we propose to exploit information about A's performance on previous benchmarks in order to warmstart its configuration on new types of benchmarks. We introduce two complementary ways in which we can exploit this information to warmstart AC methods based on a predictive model. Experiments for optimizing a very flexible modern SAT solver on twelve different instance sets show that our methods often yield substantial speedups over existing AC methods (up to 165-fold) and can also find substantially better configurations given the same compute budget.