Preference-Driven Multi-Objective Combinatorial Optimization with Conditional Computation

TL;DR

This paper introduces POCCO, a flexible framework for multi-objective combinatorial optimization that adaptively selects neural architectures for subproblems based on preferences, improving performance and generalization.

Contribution

The paper proposes a novel plug-and-play framework with conditional computation and preference-driven optimization for better solving MOCOPs.

Findings

POCCO outperforms existing methods on four benchmarks.

It demonstrates strong generalization across different problems.

Adaptive model selection improves solution quality.

Abstract

Recent deep reinforcement learning methods have achieved remarkable success in solving multi-objective combinatorial optimization problems (MOCOPs) by decomposing them into multiple subproblems, each associated with a specific weight vector. However, these methods typically treat all subproblems equally and solve them using a single model, hindering the effective exploration of the solution space and thus leading to suboptimal performance. To overcome the limitation, we propose POCCO, a novel plug-and-play framework that enables adaptive selection of model structures for subproblems, which are subsequently optimized based on preference signals rather than explicit reward values. Specifically, we design a conditional computation block that routes subproblems to specialized neural architectures. Moreover, we propose a preference-driven optimization algorithm that learns pairwise preferences between winning and losing solutions. We evaluate the efficacy and versatility of POCCO by applying it to two state-of-the-art neural methods for MOCOPs. Experimental results across four classic MOCOP benchmarks demonstrate its significant superiority and strong generalization.