Diversity By Design: Leveraging Distribution Matching for Offline Model-Based Optimization

TL;DR

This paper introduces DynAMO, a novel offline model-based optimization method that explicitly incorporates design diversity by matching the distribution of generated designs to the offline dataset, leading to more varied and high-quality solutions.

Contribution

DynAMO formulates diversity as a distribution matching problem, enabling explicit control of design diversity in offline MBO across multiple scientific domains.

Findings

DynAMO significantly improves diversity of proposed designs.

It maintains high-quality candidate discovery.

Applicable with common optimization methods.

Abstract

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.