Contrastive Representation Learning : comparemela.com

The goal of contrastive representation learning is to learn such an embedding space in which similar sample pairs stay close to each other while dissimilar ones are far apart. Contrastive learning can be applied to both supervised and unsupervised settings. When working with unsupervised data, contrastive learning is one of the most powerful approaches in self-supervised learning.
Contrastive Training Objectives In early versions of loss functions for contrastive learning, only one positive and one negative sample are involved.

Related Keywords

Prannay Khosla , Wang Isola , Salakhutdinov Hinton , Jason Wei , Ekind Cubuk , Lajanugen Logeswaran , Phillip Isola , Dmitry Kalenichenko , Tianyu Gao , Yazhe Li Oriol Vinyals , Joshua Robinson , Logeswaran Lee , Ching Yao Chuang , Monte Carlo , Geoffrey Hinton , Mathilde Caron , Florian Schroff , Iryna Gurevych , Geoff Hinton , Cutmix Yun , Nicolas Papernot , Aapo Hyv , Bohan Li , Dinghan Shen , Sumit Chopra , Wei Zou , Jianlin Su , Amoco , Zhirong Wu , Hongyi Zhang , Abe Fetterman Josh Albrecht , Nils Reimers , Facenet Schroff , Michael Gutmann , Alec Radford , James Philbin , Nicholas Frosst , Honglak Lee , Raia Hadsell , Reimers Gurevych , Yan Zhang , Kai Zou , Ashish Jaiswal , Devries Taylor , Kihyuk Sohn , Yannis Kalantidis , Ruslan Salakhutdinov , Sosuke Kobayashi , Sangdoo Yun , Tongzhou Wang , Autoaugment Cubuk , Gutmann Hyvarinen , Hongchao Fang , Randaugment Cubuk , Yann Lecun , Memory Bank , Aaron , Memoy Bank , Contrastive Self , Structured Loss , Classn Pair , Contrastive Estimation , Contrastive Predictive Coding , Nearest Neighbors Loss , Unsupervised Data Augmentation , Contrastive Hard , Barlow Twins , Bootstrap Your Own Latent , Abe Fetterman , Josh Albrecht , Swapping Assignments , Contrastive Language Image Pre Training , Contrastive Loss , Easy Data Augmentation , Contextual Augmentation , Encoder Representations , Semantic Text Similarity , Language Inference , Unified Embedding , Face Recognition , Metric Learning , Lifted Structured Feature Embedding , Nonlinear Embedding , Preserving Class Neighbourhood , Deep Metric Learning , Multi Classn Pair Loss , Improving Representations , Soft Nearest Neighbor Loss , Contrastive Representation Learning , Based Augmentation , Efficient Learning , Augmentation Policy Schedules , Barret Zoph , Beyond Empirical Risk Minimization , Regularization Strategy , Train Strong Classifiers , Localizable Features , Contrastive Self Supervised Learning , Self Supervised Learning , Redundancy Reduction , Transferable Visual Models From Natural Language , Visual Features , Contrasting Cluster Assignments , Unsupervised Learning , Contrastive Learning , Contrastive Predictive , Data Augmentation , Paradigmatic Relations , Tough To Beat Data Augmentation Approach , Natural Language Understanding , Simple Contrastive Learning , Sentence Embeddings , Hard Negative Samples , Yao Chuang , Representation Learning , Long Read , Language Model ,