A Decoupled Basis-Vector-Driven Generative Framework for Dynamic Multi-Objective Optimization
This work addresses dynamic multi-objective optimization problems, which are important for real-world applications like robotics and scheduling, by providing a more efficient and accurate solution method.
The paper tackles dynamic multi-objective optimization by proposing DB-GEN, a framework that addresses challenges like non-linear coupling, negative transfer, and cold-start problems through wavelet decomposition, sparse dictionary learning, and surrogate-assisted search. It achieves improved tracking accuracy with zero-shot generation in approximately 0.2 seconds per environmental change, pre-trained on 120 million solutions.
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and data sparsity, primarily facing three challenges: the non-linear coupling of dynamic modes, negative transfer from outdated historical data, and the cold-start problem during environmental switches.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.