Implicit Contrastive Representation Learning with Guided Stop-gradient

• URL: http://arxiv.org/abs/2503.09058v1
• Date: Wed, 12 Mar 2025 04:46:53 GMT
• Title: Implicit Contrastive Representation Learning with Guided Stop-gradient
• Authors: Byeongchan Lee, Sehyun Lee,
• Abstract summary: We introduce a methodology to implicitly incorporate the idea of contrastive learning.<n>We show that our method stabilizes training and boosts performance.<n>The algorithms with our method work well with small batch sizes and do not collapse even when there is no predictor.
• Score: 0.0
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: 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 into a degenerate solution. To address the issue, in contrastive learning, a contrastive loss is used to prevent collapse by moving representations of negative pairs away from each other. But it is known that algorithms with negative sampling are not robust to a reduction in the number of negative samples. So, on the other hand, there are algorithms that do not use negative pairs. Many positive-only algorithms adopt asymmetric network architecture consisting of source and target encoders as a key factor in coping with collapse. By exploiting the asymmetric architecture, we introduce a methodology to implicitly incorporate the idea of contrastive learning. As its implementation, we present a novel method guided stop-gradient. We apply our method to benchmark algorithms SimSiam and BYOL and show that our method stabilizes training and boosts performance. We also show that the algorithms with our method work well with small batch sizes and do not collapse even when there is no predictor. The code is available at https://github.com/bych-lee/gsg.

Related papers

• Understanding Collapse in Non-Contrastive Learning [122.2499276246997]
  We show that 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.
  arXiv  Detail & Related papers  (2022-09-29T17:59:55Z)
• Non-contrastive representation learning for intervals from well logs [58.70164460091879]
  The representation learning problem in the oil & gas industry aims to construct a model that provides a representation based on logging data for a well interval. One of the possible approaches is self-supervised learning (SSL) We are the first to introduce non-contrastive SSL for well-logging data.
  arXiv  Detail & Related papers  (2022-09-28T13:27:10Z)
• Chaos is a Ladder: A New Theoretical Understanding of Contrastive Learning via Augmentation Overlap [64.60460828425502]
  We propose a new guarantee on the downstream performance of contrastive learning. Our new theory hinges on the insight that the support of different intra-class samples will become more overlapped under aggressive data augmentations. We propose an unsupervised model selection metric ARC that aligns well with downstream accuracy.
  arXiv  Detail & Related papers  (2022-03-25T05:36:26Z)
• Exploring the Equivalence of Siamese Self-Supervised Learning via A Unified Gradient Framework [43.76337849044254]
  Self-supervised learning has shown its great potential to extract powerful visual representations without human annotations. Various works are proposed to deal with self-supervised learning from different perspectives. We propose UniGrad, a simple but effective gradient form for self-supervised learning.
  arXiv  Detail & Related papers  (2021-12-09T18:59:57Z)
• MIO : Mutual Information Optimization using Self-Supervised Binary Contrastive Learning [19.5917119072985]
  We model contrastive learning into a binary classification problem to predict if a pair is positive or not. The proposed method outperforms the state-of-the-art algorithms on benchmark datasets like STL-10, CIFAR-10, CIFAR-100.
  arXiv  Detail & Related papers  (2021-11-24T17:51:29Z)
• Simple Stochastic and Online Gradient DescentAlgorithms for Pairwise Learning [65.54757265434465]
  Pairwise learning refers to learning tasks where the loss function depends on a pair instances. Online descent (OGD) is a popular approach to handle streaming data in pairwise learning. In this paper, we propose simple and online descent to methods for pairwise learning.
  arXiv  Detail & Related papers  (2021-11-23T18:10:48Z)
• Trip-ROMA: Self-Supervised Learning with Triplets and Random Mappings [59.32440962369532]
  We show that a simple Triplet-based loss can achieve surprisingly good performance without requiring large batches or asymmetry designs. To alleviate the over-fitting problem in small data regimes, we propose a simple plug-and-play RandOm MApping (ROMA) strategy.
  arXiv  Detail & Related papers  (2021-07-22T02:06:38Z)
• Understanding self-supervised Learning Dynamics without Contrastive Pairs [72.1743263777693]
  Contrastive approaches to self-supervised learning (SSL) learn representations by minimizing the distance between two augmented views of the same data point. BYOL and SimSiam, show remarkable performance it without negative pairs. We study the nonlinear learning dynamics of non-contrastive SSL in simple linear networks.
  arXiv  Detail & Related papers  (2021-02-12T22:57:28Z)
• LoCo: Local Contrastive Representation Learning [93.98029899866866]
  We show that by overlapping local blocks stacking on top of each other, we effectively increase the decoder depth and allow upper blocks to implicitly send feedbacks to lower blocks. This simple design closes the performance gap between local learning and end-to-end contrastive learning algorithms for the first time.
  arXiv  Detail & Related papers  (2020-08-04T05:41:29Z)
• DeepMP for Non-Negative Sparse Decomposition [14.790515227906257]
  Non-negative signals form an important class of sparse signals. greedy and convex relaxed algorithms are among the most popular methods. One such modification has been proposed for Matching Pursuit (MP) based algorithms.
  arXiv  Detail & Related papers  (2020-07-28T14:52:06Z)
• Whitening for Self-Supervised Representation Learning [129.57407186848917]
  We propose a new loss function for self-supervised representation learning (SSL) based on the whitening of latent-space features. Our solution does not require asymmetric networks and it is conceptually simple.
  arXiv  Detail & Related papers  (2020-07-13T12:33:25Z)
• SCE: Scalable Network Embedding from Sparsest Cut [20.08464038805681]
  Large-scale network embedding is to learn a latent representation for each node in an unsupervised manner. A key of success to such contrastive learning methods is how to draw positive and negative samples. In this paper, we propose SCE for unsupervised network embedding only using negative samples for training.
  arXiv  Detail & Related papers  (2020-06-30T03:18:15Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.