Sharpness-Aware Black-Box Optimization

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Ye, Feiyang, Lyu, Yueming, Wang, Xuehao, Sugiyama, Masashi, Zhang, Yu, Tsang, Ivor
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866917804962217984
author Ye, Feiyang
Lyu, Yueming
Wang, Xuehao
Sugiyama, Masashi
Zhang, Yu
Tsang, Ivor
author_facet Ye, Feiyang
Lyu, Yueming
Wang, Xuehao
Sugiyama, Masashi
Zhang, Yu
Tsang, Ivor
contents Black-box optimization algorithms have been widely used in various machine learning problems, including reinforcement learning and prompt fine-tuning. However, directly optimizing the training loss value, as commonly done in existing black-box optimization methods, could lead to suboptimal model quality and generalization performance. To address those problems in black-box optimization, we propose a novel Sharpness-Aware Black-box Optimization (SABO) algorithm, which applies a sharpness-aware minimization strategy to improve the model generalization. Specifically, the proposed SABO method first reparameterizes the objective function by its expectation over a Gaussian distribution. Then it iteratively updates the parameterized distribution by approximated stochastic gradients of the maximum objective value within a small neighborhood around the current solution in the Gaussian distribution space. Theoretically, we prove the convergence rate and generalization bound of the proposed SABO algorithm. Empirically, extensive experiments on the black-box prompt fine-tuning tasks demonstrate the effectiveness of the proposed SABO method in improving model generalization performance.
format Preprint
id arxiv_https___arxiv_org_abs_2410_12457
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Sharpness-Aware Black-Box Optimization
Ye, Feiyang
Lyu, Yueming
Wang, Xuehao
Sugiyama, Masashi
Zhang, Yu
Tsang, Ivor
Machine Learning
Artificial Intelligence
Black-box optimization algorithms have been widely used in various machine learning problems, including reinforcement learning and prompt fine-tuning. However, directly optimizing the training loss value, as commonly done in existing black-box optimization methods, could lead to suboptimal model quality and generalization performance. To address those problems in black-box optimization, we propose a novel Sharpness-Aware Black-box Optimization (SABO) algorithm, which applies a sharpness-aware minimization strategy to improve the model generalization. Specifically, the proposed SABO method first reparameterizes the objective function by its expectation over a Gaussian distribution. Then it iteratively updates the parameterized distribution by approximated stochastic gradients of the maximum objective value within a small neighborhood around the current solution in the Gaussian distribution space. Theoretically, we prove the convergence rate and generalization bound of the proposed SABO algorithm. Empirically, extensive experiments on the black-box prompt fine-tuning tasks demonstrate the effectiveness of the proposed SABO method in improving model generalization performance.
title Sharpness-Aware Black-Box Optimization
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2410.12457