paper-with-me

Papers

A Theoretical Characterization of Optimal Data Augmentations in Self-Supervised Learning

2024-11-04 · Shlomo Libo Feigin, Maximilian Fleissner, Debarghya Ghoshdastidar

Data augmentations play an important role in the recent success of Self-Supervised Learning (SSL). While commonly viewed as encoding invariances into the learned representations, this interpretation overlooks the impact of the pretraining architecture and suggests that SSL would require diverse augmentations which resemble the data to work well. However, these assumptions do not align with empirical evidence, encouraging further theoretical understanding to guide the principled design of augmentations in new domains. To this end, we use kernel theory to derive analytical expressions for data augmentations that achieve desired target representations after pretraining. We consider two popular non-contrastive losses, VICReg and Barlow Twins, and provide an algorithm to construct such augmentations. Our analysis shows that augmentations need not be similar to the data to learn useful representations, nor be diverse, and that the architecture has a significant impact on the optimal augmentations.

📄 PDF Abstract BibTeX arXiv:2411.01767

Code (0)

등록된 구현이 없습니다.

Tasks

Self-Supervised Learning

Methods 이 논문이 사용한 방법론

ALIGN In the ALIGN method, visual and language representations are jointly trained from noisy image alt-text data. The image and text encoders are learned via contrastive loss…
Barlow Twins Barlow Twins is a self-supervised learning method that applies redundancy-reduction — a principle first proposed in neuroscience — to self supervised learning. The objective…

Similar Papers 제목 키워드 기반

Understanding Augmentation-based Self-Supervised Representation Learning via RKHS Approximation and Regression

2023-06-01 · Runtian Zhai, Bingbin Liu, Andrej Risteski, Zico Kolter 외

Data augmentation is critical to the empirical success of modern self-supervised representation learning, such as contrastive learning and masked language modeling. However, a theoretical understanding of the exact role …

Contrastive LearningData AugmentationGeneralization BoundsLanguage Modeling+6

An Analysis of Self-supervised Pre-training with Dependent Samples

2026-09-04 · Maximilian Fleissner, Debarghya Ghoshdastidar, Samory Kpotufe arxiv

Self-supervised learning relies on so-called data augmentations $φ(x)$ of unlabeled datapoints $x$ --- for example, masking random pixels in an image $x$ --- that should leave the label of $x$ invariant and are often use…

Self-Supervised Learning

Representation Learning via Invariant Causal Mechanisms

2020-10-15 · Jovana Mitrovic, Brian McWilliams, Jacob Walker, Lars Buesing 외

Self-supervised learning has emerged as a strategy to reduce the reliance on costly supervised signal by pretraining representations only using unlabeled data. These methods combine heuristic proxy classification tasks w…

Contrastive LearningOut-of-Distribution GeneralizationRepresentation LearningSelf-Supervised Image Classification+1

Homomorphic Self-Supervised Learning

2022-11-15 · T. Anderson Keller, Xavier Suau, Luca Zappella

In this work, we observe that many existing self-supervised learning algorithms can be both unified and generalized when seen through the lens of equivariant representations. Specifically, we introduce a general framewor…

Self-Supervised Learning

Structuring Representation Geometry with Rotationally Equivariant Contrastive Learning

2023-06-24 · Sharut Gupta, Joshua Robinson, Derek Lim, Soledad Villar 외

Self-supervised learning converts raw perceptual data such as images to a compact space where simple Euclidean distances measure meaningful variations in data. In this paper, we extend this formulation by adding addition…

Contrastive LearningSelf-Supervised Learning