Swift Sampler: Efficient Learning of Sampler by 10 Parameters

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Yao, Jiawei, Li, Chuming, Xiao, Canran
Format: Preprint
Publié: 2024
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866912062956896256
author Yao, Jiawei
Li, Chuming
Xiao, Canran
author_facet Yao, Jiawei
Li, Chuming
Xiao, Canran
contents Data selection is essential for training deep learning models. An effective data sampler assigns proper sampling probability for training data and helps the model converge to a good local minimum with high performance. Previous studies in data sampling are mainly based on heuristic rules or learning through a huge amount of time-consuming trials. In this paper, we propose an automatic \textbf{swift sampler} search algorithm, \textbf{SS}, to explore automatically learning effective samplers efficiently. In particular, \textbf{SS} utilizes a novel formulation to map a sampler to a low dimension of hyper-parameters and uses an approximated local minimum to quickly examine the quality of a sampler. Benefiting from its low computational expense, \textbf{SS} can be applied on large-scale data sets with high efficiency. Comprehensive experiments on various tasks demonstrate that \textbf{SS} powered sampling can achieve obvious improvements (e.g., 1.5\% on ImageNet) and transfer among different neural networks. Project page: https://github.com/Alexander-Yao/Swift-Sampler.
format Preprint
id arxiv_https___arxiv_org_abs_2410_05578
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Swift Sampler: Efficient Learning of Sampler by 10 Parameters
Yao, Jiawei
Li, Chuming
Xiao, Canran
Machine Learning
Artificial Intelligence
Data selection is essential for training deep learning models. An effective data sampler assigns proper sampling probability for training data and helps the model converge to a good local minimum with high performance. Previous studies in data sampling are mainly based on heuristic rules or learning through a huge amount of time-consuming trials. In this paper, we propose an automatic \textbf{swift sampler} search algorithm, \textbf{SS}, to explore automatically learning effective samplers efficiently. In particular, \textbf{SS} utilizes a novel formulation to map a sampler to a low dimension of hyper-parameters and uses an approximated local minimum to quickly examine the quality of a sampler. Benefiting from its low computational expense, \textbf{SS} can be applied on large-scale data sets with high efficiency. Comprehensive experiments on various tasks demonstrate that \textbf{SS} powered sampling can achieve obvious improvements (e.g., 1.5\% on ImageNet) and transfer among different neural networks. Project page: https://github.com/Alexander-Yao/Swift-Sampler.
title Swift Sampler: Efficient Learning of Sampler by 10 Parameters
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2410.05578