A Simple Data Mixing Prior for Improving Self-Supervised Learning

06/15/2022
by   Sucheng Ren, et al.
19

Data mixing (e.g., Mixup, Cutmix, ResizeMix) is an essential component for advancing recognition models. In this paper, we focus on studying its effectiveness in the self-supervised setting. By noticing the mixed images that share the same source images are intrinsically related to each other, we hereby propose SDMP, short for Simple Data Mixing Prior, to capture this straightforward yet essential prior, and position such mixed images as additional positive pairs to facilitate self-supervised representation learning. Our experiments verify that the proposed SDMP enables data mixing to help a set of self-supervised learning frameworks (e.g., MoCo) achieve better accuracy and out-of-distribution robustness. More notably, our SDMP is the first method that successfully leverages data mixing to improve (rather than hurt) the performance of Vision Transformers in the self-supervised setting. Code is publicly available at https://github.com/OliverRensu/SDMP

READ FULL TEXT

page 1

page 3

research
08/24/2020

Self-Supervised Learning for Large-Scale Unsupervised Image Clustering

Unsupervised learning has always been appealing to machine learning rese...
research
10/11/2022

OPERA: Omni-Supervised Representation Learning with Hierarchical Supervisions

The pretrain-finetune paradigm in modern computer vision facilitates the...
research
08/03/2021

Solo-learn: A Library of Self-supervised Methods for Visual Representation Learning

This paper presents solo-learn, a library of self-supervised methods for...
research
10/25/2021

Self-supervised similarity search for large scientific datasets

We present the use of self-supervised learning to explore and exploit la...
research
09/30/2021

Mining for strong gravitational lenses with self-supervised learning

We employ self-supervised representation learning to distill information...
research
02/06/2023

Trust, but Verify: Using Self-Supervised Probing to Improve Trustworthiness

Trustworthy machine learning is of primary importance to the practical d...
research
12/26/2022

SMMix: Self-Motivated Image Mixing for Vision Transformers

CutMix is a vital augmentation strategy that determines the performance ...

Please sign up or login with your details

Forgot password? Click here to reset