SASSL: Enhancing Self-Supervised Learning via Neural Style Transfer

Renan A. Rojas-Gomez, Karan Singhal, Ali Etemad, Alex Bijamov, Warren Richard Morningstar, Philip Andrew Mansfield

Abstract: Existing data augmentation in self-supervised learning, while diverse, fails to preserve the inherent structure of natural images. This results in distorted augmented samples with compromised semantic information, ultimately impacting downstream performance. To overcome this limitation, we propose SASSL: Style Augmentations for Self Supervised Learning, a novel data augmentation technique based on Neural Style Transfer. SASSL decouples semantic and stylistic attributes in images and applies transformations exclusively to their style while preserving content, generating diverse samples that better retain semantic information. SASSL boosts top-1 image classification accuracy on ImageNet by up to 2 percentage points compared to established self-supervised methods like MoCo, SimCLR, and BYOL, while achieving superior transfer learning performance across various datasets. Because SASSL can be performed asynchronously as part of the data augmentation pipeline, these performance impacts can be obtained with no change in pretraining throughput.

Submission Length: Regular submission (no more than 12 pages of main content)

Assigned Action Editor: ~Nicolas_THOME2

Submission Number: 2789