NeuroPareto: Calibrated Acquisition for Costly Many-Goal Search in Vast Parameter Spaces
This addresses the problem of costly optimization for researchers and practitioners in fields like engineering, though it appears incremental as it builds on existing classifier-enhanced and surrogate-assisted methods.
The paper tackled the challenge of multi-objective optimization in high-dimensional spaces under computational constraints by developing NeuroPareto, which integrates filtering, uncertainty disentanglement, and acquisition strategies to efficiently generate high-quality candidates, showing consistent outperformance in Pareto proximity and hypervolume on benchmark tasks.
The pursuit of optimal trade-offs in high-dimensional search spaces under stringent computational constraints poses a fundamental challenge for contemporary multi-objective optimization. We develop NeuroPareto, a cohesive architecture that integrates rank-centric filtering, uncertainty disentanglement, and history-conditioned acquisition strategies to navigate complex objective landscapes. A calibrated Bayesian classifier estimates epistemic uncertainty across non-domination tiers, enabling rapid generation of high-quality candidates with minimal evaluation cost. Deep Gaussian Process surrogates further separate predictive uncertainty into reducible and irreducible components, providing refined predictive means and risk-aware signals for downstream selection. A lightweight acquisition network, trained online from historical hypervolume improvements, guides expensive evaluations toward regions balancing convergence and diversity. With hierarchical screening and amortized surrogate updates, the method maintains accuracy while keeping computational overhead low. Experiments on DTLZ and ZDT suites and a subsurface energy extraction task show that NeuroPareto consistently outperforms classifier-enhanced and surrogate-assisted baselines in Pareto proximity and hypervolume.