让离线优化生成更多样设计，提升创新可能性。
Diversity By Design: Leveraging Distribution Matching for Offline Model-Based Optimization
- 将多样性设为显式目标，通过分布匹配实现
- 在多领域实验中显著提升设计多样性，同时保持高质量
- 适合需要多样解的科研与工程优化场景
离线模型基于优化（MBO）的目标是在仅有离线数据集的情况下，提出能最大化奖励函数的新设计。然而一个重要需求是生成一组多样化的最终候选设计，以覆盖多种最优和近优配置。本文提出动态对抗模型优化（DynAMO），作为一种新方法，将设计多样性作为显式目标融入任意MBO问题。核心思想是将多样性建模为分布匹配问题，使生成设计的分布匹配离线数据集中蕴含的固有多样性。在多个科学领域的广泛实验表明，DynAMO可与常见优化方法结合，显著提升所提设计的多样性，同时仍能发现高质量候选。
原文摘要 · Abstract (English)
The goal of offline model-based optimization (MBO) is to propose new designs that maximize a reward function given only an offline dataset. However, an important desiderata is to also propose a diverse set of final candidates that capture many optimal and near-optimal design configurations. We propose Diversity in Adversarial Model-based Optimization (DynAMO) as a novel method to introduce design diversity as an explicit objective into any MBO problem. Our key insight is to formulate diversity as a distribution matching problem where the distribution of generated designs captures the inherent diversity contained within the offline dataset. Extensive experiments spanning multiple scientific domains show that DynAMO can be used with common optimization methods to significantly improve the diversity of proposed designs while still discovering high-quality candidates.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。