SANE: Strategic Autonomous Non-Smooth Exploration for Multiple Optima Discovery in Multi-modal and Non-differentiable Black-box Functions

TL;DR

SANE is a novel Bayesian optimization method designed to efficiently explore multi-modal, non-differentiable black-box functions by finding multiple optima, integrating human knowledge to distinguish true from false regions, and outperforming classical BO in noisy experimental data.

Contribution

This work introduces SANE, a strategic exploration algorithm that effectively discovers multiple optima in complex, noisy black-box functions, incorporating human-driven surrogate gating for improved accuracy.

Findings

SANE outperforms classical Bayesian optimization in identifying multiple optima.

SANE effectively handles noisy, multi-modal, non-differentiable functions.

Application to real experimental data demonstrates higher coverage of scientific discovery.

Abstract

Both computational and experimental material discovery bring forth the challenge of exploring multidimensional and multimodal parameter spaces, such as phase diagrams of Hamiltonians with multiple interactions, composition spaces of combinatorial libraries, material structure image spaces, and molecular embedding spaces. Often these systems are black-box and time-consuming to evaluate, which resulted in strong interest towards active learning methods such as Bayesian optimization (BO). However, these systems are often noisy which make the black box function severely multi-modal and non-differentiable, where a vanilla BO can get overly focused near a single or faux optimum, deviating from the broader goal of scientific discovery. To address these limitations, here we developed Strategic Autonomous Non-Smooth Exploration (SANE) to facilitate an intelligent Bayesian optimized navigation with a proposed cost-driven probabilistic acquisition function to find multiple global and local optimal regions, avoiding the tendency to becoming trapped in a single optimum. To distinguish between a true and false optimal region due to noisy experimental measurements, a human (domain) knowledge driven dynamic surrogate gate is integrated with SANE. We implemented the gate-SANE into a pre-acquired Piezoresponse spectroscopy data of a ferroelectric combinatorial library with high noise levels in specific regions, and a piezoresponse force microscopy (PFM) hyperspectral data. SANE demonstrated better performance than classical BO to facilitate the exploration of multiple optimal regions and thereby prioritized learning with higher coverage of scientific values in autonomous experiments. Our work showcases the potential application of this method to real-world experiment, where such combined strategic and human intervening approaches can be critical to unlocking new discoveries in autonomous research.