Preference-Driven Multi-Objective Combinatorial Optimization with Conditional Computation
This addresses a bottleneck in multi-objective combinatorial optimization for researchers and practitioners, offering an incremental improvement over existing neural methods.
The paper tackled the suboptimal performance of deep reinforcement learning methods in multi-objective combinatorial optimization by proposing POCCO, a framework that adaptively selects model structures for subproblems and uses preference-driven optimization, achieving significant superiority and strong generalization across four benchmarks.
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.