paper-with-me

홈 › Papers

Understanding Collapse in Non-Contrastive Siamese Representation Learning

2022-09-29 · Alexander C. Li, Alexei A. Efros, Deepak Pathak

Contrastive methods have led a recent surge in the performance of self-supervised representation learning (SSL). Recent methods like BYOL or SimSiam purportedly distill these contrastive methods down to their essence, removing bells and whistles, including the negative examples, that do not contribute to downstream performance. These "non-contrastive" methods work surprisingly well without using negatives even though the global minimum lies at trivial collapse. We empirically analyze these non-contrastive methods and find that SimSiam is extraordinarily sensitive to dataset and model size. In particular, SimSiam representations undergo partial dimensional collapse if the model is too small relative to the dataset size. We propose a metric to measure the degree of this collapse and show that it can be used to forecast the downstream task performance without any fine-tuning or labels. We further analyze architectural design choices and their effect on the downstream performance. Finally, we demonstrate that shifting to a continual learning setting acts as a regularizer and prevents collapse, and a hybrid between continual and multi-epoch training can improve linear probe accuracy by as many as 18 percentage points using ResNet-18 on ImageNet. Our project page is at https://alexanderli.com/noncontrastive-ssl/.

📄 PDF Abstract BibTeX arXiv:2209.15007

Code (1)

alexlioralexli/noncontrastive-ssl 공식 구현 pytorch

Tasks

Continual LearningContrastive LearningRepresentation Learning

Methods 이 논문이 사용한 방법론

BYOL 설명 없음

Similar Papers 제목 키워드 기반

How Does SimSiam Avoid Collapse Without Negative Samples? A Unified Understanding with Self-supervised Contrastive Learning

2022-03-30 · Chaoning Zhang, Kang Zhang, Chenshuang Zhang, Trung X. Pham 외

To avoid collapse in self-supervised learning (SSL), a contrastive loss is widely used but often requires a large number of negative samples. Without negative samples yet achieving competitive performance, a recent work …

Contrastive LearningSelf-Supervised Learning

How Does SimSiam Avoid Collapse Without Negative Samples? Towards a Unified Understanding of Progress in SSL

2021-09-29 · ICLR 2022 4 · Chaoning Zhang, Kang Zhang, Chenshuang Zhang, Trung X. Pham 외

Towards avoiding collapse in self-supervised learning (SSL), contrastive loss is widely used but often requires a large number of negative samples. Without negative samples yet achieving competitive performance one recen…

Self-Supervised Learning

Implicit Contrastive Representation Learning with Guided Stop-gradient

2025-03-12 · NeurIPS 2023 11 · Byeongchan Lee, Sehyun Lee

In self-supervised representation learning, Siamese networks are a natural architecture for learning transformation-invariance by bringing representations of positive pairs closer together. But it is prone to collapse in…

Contrastive LearningRepresentation Learning

Understanding Dimensional Collapse in Contrastive Self-supervised Learning

2021-10-18 · ICLR 2022 4 · Li Jing, Pascal Vincent, Yann Lecun, Yuandong Tian

Self-supervised visual representation learning aims to learn useful representations without relying on human annotations. Joint embedding approach bases on maximizing the agreement between embedding vectors from differen…

Contrastive LearningLearning TheoryRepresentation LearningSelf-Supervised Learning

SupSiam: Non-contrastive Auxiliary Loss for Learning from Molecular Conformers

2023-02-15 · Michael Maser, Ji Won Park, Joshua Yao-Yu Lin, Jae Hyeon Lee 외

We investigate Siamese networks for learning related embeddings for augmented samples of molecular conformers. We find that a non-contrastive (positive-pair only) auxiliary task aids in supervised training of Euclidean n…

Activity PredictionDrug Discoveryregression