Saved in:
Bibliographic Details
Main Authors: Jiao, Yuling, Ma, Wensen, Sun, Defeng, Wang, Hansheng, Wang, Yang
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2502.14424
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912460895682560
author Jiao, Yuling
Ma, Wensen
Sun, Defeng
Wang, Hansheng
Wang, Yang
author_facet Jiao, Yuling
Ma, Wensen
Sun, Defeng
Wang, Hansheng
Wang, Yang
contents In this paper, we propose a novel self-supervised transfer learning method called \underline{\textbf{D}}istribution \underline{\textbf{M}}atching (DM), which drives the representation distribution toward a predefined reference distribution while preserving augmentation invariance. DM results in a learned representation space that is intuitively structured and therefore easy to interpret. Experimental results across multiple real-world datasets and evaluation metrics demonstrate that DM performs competitively on target classification tasks compared to existing self-supervised transfer learning methods. Additionally, we provide robust theoretical guarantees for DM, including a population theorem and an end-to-end sample theorem. The population theorem bridges the gap between the self-supervised learning task and target classification accuracy, while the sample theorem shows that, even with a limited number of samples from the target domain, DM can deliver exceptional classification performance, provided the unlabeled sample size is sufficiently large.
format Preprint
id arxiv_https___arxiv_org_abs_2502_14424
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Distribution Matching for Self-Supervised Transfer Learning
Jiao, Yuling
Ma, Wensen
Sun, Defeng
Wang, Hansheng
Wang, Yang
Machine Learning
Artificial Intelligence
Methodology
In this paper, we propose a novel self-supervised transfer learning method called \underline{\textbf{D}}istribution \underline{\textbf{M}}atching (DM), which drives the representation distribution toward a predefined reference distribution while preserving augmentation invariance. DM results in a learned representation space that is intuitively structured and therefore easy to interpret. Experimental results across multiple real-world datasets and evaluation metrics demonstrate that DM performs competitively on target classification tasks compared to existing self-supervised transfer learning methods. Additionally, we provide robust theoretical guarantees for DM, including a population theorem and an end-to-end sample theorem. The population theorem bridges the gap between the self-supervised learning task and target classification accuracy, while the sample theorem shows that, even with a limited number of samples from the target domain, DM can deliver exceptional classification performance, provided the unlabeled sample size is sufficiently large.
title Distribution Matching for Self-Supervised Transfer Learning
topic Machine Learning
Artificial Intelligence
Methodology
url https://arxiv.org/abs/2502.14424