NeuroPareto: Calibrated Acquisition for Costly Many-Goal Search in Vast Parameter Spaces
cs.LG, cs.NE
Submitted: 2026-02-03
Updated: 2026-10-29
Comments: 39 pages, 19 figures
License: http://creativecommons.org/licenses/by/4.0/
The gist: The pursuit of optimal trade-offs in high-dimensional search spaces under stringent computational constraints poses a fundamental challenge for contemporary multi-objective optimization.
Terminology
Abstract
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.
Sources
- Rank-Based Learning and Local Model Based Evolutionary Algorithm for High-Dimensional Expensive Multi-Objective Problems
- Bayesian Optimization of Function Networks with Partial Evaluations
- Enhancing Monte Carlo Dropout Performance for Uncertainty Quantification
- Amortized Variational Inference for Deep Gaussian Processes
- Sparse Gaussian Neural Processes
- Permutation-based multi-objective evolutionary feature selection for high-dimensional data
- RCBSF: A Multi-Agent Framework for Automated Contract Revision via Stackelberg Game
Related papers
- A Physics-Regulated Neural Framework for Learning 3D Grain Growth Dynamics
- FlexMoE: Learning One-for-All Multi-Budget Expert Pruning for Flexible MoE Deployment
- OptiMind: Improving Small Language Models for Optimization Modeling through Formulation Guidance and Multi-Turn Learning
- Staying in the Sweet Spot: Boost Reasoning Evolution via Responsive Rollout Difficulty Manipulation
- Examining the Impact of Data and Training Methods on Emergent and Subliminal Misalignment
- Rethinking Black-Box On-Policy Distillation: A Rubric-based Analysis