paper-with-me

Papers

Conditional Negative Sampling for Contrastive Learning of Visual Representations

2020-10-05 · ICLR 2021 1 · Mike Wu, Milan Mosse, Chengxu Zhuang, Daniel Yamins, Noah Goodman

Recent methods for learning unsupervised visual representations, dubbed contrastive learning, optimize the noise-contrastive estimation (NCE) bound on mutual information between two views of an image. NCE uses randomly sampled negative examples to normalize the objective. In this paper, we show that choosing difficult negatives, or those more similar to the current instance, can yield stronger representations. To do this, we introduce a family of mutual information estimators that sample negatives conditionally -- in a "ring" around each positive. We prove that these estimators lower-bound mutual information, with higher bias but lower variance than NCE. Experimentally, we find our approach, applied on top of existing models (IR, CMC, and MoCo) improves accuracy by 2-5% points in each case, measured by linear evaluation on four standard image datasets. Moreover, we find continued benefits when transferring features to a variety of new image distributions from the Meta-Dataset collection and to a variety of downstream tasks such as object detection, instance segmentation, and keypoint detection.

📄 PDF Abstract BibTeX arXiv:2010.02037

Code (1)

mhw32/contrastive-ring-public pytorch

Tasks

Contrastive LearningInstance SegmentationKeypoint DetectionLinear evaluationobject-detectionObject DetectionSemantic Segmentation

Similar Papers 제목 키워드 기반

Conditional Contrastive Learning with Kernel

2022-02-11 · ICLR 2022 4 · Yao-Hung Hubert Tsai, Tianqin Li, Martin Q. Ma, Han Zhao 외

Conditional contrastive learning frameworks consider the conditional sampling procedure that constructs positive or negative data pairs conditioned on specific variables. Fair contrastive learning constructs negative pai…

Contrastive Learning

Conditional Contrastive Learning for Improving Fairness in Self-Supervised Learning

2021-06-05 · Martin Q. Ma, Yao-Hung Hubert Tsai, Paul Pu Liang, Han Zhao 외

Contrastive self-supervised learning (SSL) learns an embedding space that maps similar data pairs closer and dissimilar data pairs farther apart. Despite its success, one issue has been overlooked: the fairness aspect of…

AttributeContrastive LearningFairnessRepresentation Learning+1

Enhancing Multimodal Entity Linking with Jaccard Distance-based Conditional Contrastive Learning and Contextual Visual Augmentation

2025-01-24 · Cong-Duy Nguyen, Xiaobao Wu, Thong Nguyen, Shuai Zhao 외

Previous research on multimodal entity linking (MEL) has primarily employed contrastive learning as the primary objective. However, using the rest of the batch as negative samples without careful consideration, these stu…

Contrastive LearningEntity Linking

Integrating Prior Knowledge in Contrastive Learning with Kernel

2022-06-03 · Benoit Dufumier, Carlo Alberto Barbano, Robin Louiset, Edouard Duchesnay 외

Data augmentation is a crucial component in unsupervised contrastive learning (CL). It determines how positive samples are defined and, ultimately, the quality of the learned representation. In this work, we open the doo…

Contrastive LearningData Augmentation

Active Contrastive Learning of Audio-Visual Video Representations

2020-08-31 · ICLR 2021 1 · Shuang Ma, Zhaoyang Zeng, Daniel McDuff, Yale Song

Contrastive learning has been shown to produce generalizable representations of audio and visual data by maximizing the lower bound on the mutual information (MI) between different views of an instance. However, obtainin…

Contrastive LearningRepresentation LearningVideo Classification