A Decoupled Basis-Vector-Driven Generative Framework for Dynamic Multi-Objective Optimization

TL;DR

This paper introduces DB-GEN, a novel generative framework for dynamic multi-objective optimization that effectively tracks moving Pareto fronts by decoupling evolutionary modes, learning transferable bases, and enabling zero-shot inference.

Contribution

The paper presents a decoupled basis-vector-driven generative framework that addresses key challenges in dynamic optimization, including non-linear coupling, negative transfer, and cold-start issues, with a scalable zero-shot approach.

Findings

DB-GEN achieves superior tracking accuracy on dynamic benchmarks.

It performs online inference in milliseconds without retraining.

The framework effectively mitigates negative transfer and cold-start problems.

Abstract

Dynamic multi-objective optimization requires continuous tracking of moving Pareto fronts. Existing methods struggle with irregular mutations and data sparsity, primarily facing three challenges: the non-linear coupling of dynamic modes, negative transfer from outdated historical data, and the cold-start problem during environmental switches. To address these issues, this paper proposes a decoupled basis-vector-driven generative framework (DB-GEN). First, to resolve non-linear coupling, the framework employs the discrete wavelet transform to separate evolutionary trajectories into low-frequency trends and high-frequency details. Second, to mitigate negative transfer, it learns transferable basis vectors via sparse dictionary learning rather than directly memorizing historical instances. Recomposing these bases under a topology-aware contrastive constraint constructs a structured latent manifold. Finally, to overcome the cold-start problem, a surrogate-assisted search paradigm samples initial populations from this manifold. Pre-trained on 120 million solutions, DB-GEN performs direct online inference without retraining or fine-tuning. This zero-shot generation process executes in milliseconds, requiring approximately 0.2 seconds per environmental change. Experimental results demonstrate that DB-GEN improves tracking accuracy across various dynamic benchmarks compared to existing algorithms.