Momentum^2 Teacher: Momentum Teacher with Momentum Statistics for Self-Supervised Learning
Not on the map. Why?
Abstract (arXiv)
In this paper, we present a novel approach, Momentum2 Teacher, for student-teacher based self-supervised learning. The approach performs momentum update on both network weights and batch normalization (BN) statistics. The teacher's weight is a momentum update of the student, and the teacher's BN statistics is a momentum update of those in history. The Momentum2 Teacher is simple and efficient. It can achieve the state of the art results (74.5%) under ImageNet linear evaluation protocol using small-batch size(eg, 128), without requiring large-batch training on special hardware like TPU or inefficient across GPU operation (eg, shuffling BN, synced BN). Our implementation and pre-trained models will be given on GitHubfootnotehttps://github.com/zengarden/momentum2-teacher.
Key facts. Cited by 14 papers in the map (data through 14 Oct 2026). Builds on 53 of its 64 references that are in the map. First posted on arXiv 19 Jan 2021.
Builds on
53 of 64 references.
References · 53
- 2015Deep Residual Learning for Image Recognitioncited by 52,096
- 2014Microsoft COCO: Common Objects in Contextcited by 17,608
- 2020A Simple Framework for Contrastive Learning of Visual Representationscited by 12,049
- 2018Representation Learning with Contrastive Predictive Codingcited by 7,777
- 2019Momentum Contrast for Unsupervised Visual Representation Learningcited by 7,513
- 2017Mask R-CNNcited by 7,277
- 2015Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shiftcited by 7,005
- 2016Layer Normalizationcited by 5,669
- 2016SGDR: Stochastic Gradient Descent with Warm Restartscited by 5,045
- 2016Feature Pyramid Networks for Object Detectioncited by 5,036
- 2020Bootstrap your own latent: A new approach to self-supervised Learningcited by 4,750
- 2020Exploring Simple Siamese Representation Learningcited by 2,866
38 more
- 2020Unsupervised Learning of Visual Features by Contrasting Cluster Assignmentscited by 2,857
- 2020Improved Baselines with Momentum Contrastive Learningcited by 2,287
- 2020FixMatch: Simplifying Semi-Supervised Learning with Consistency and Confidencecited by 2,085
- 2018Unsupervised Feature Learning via Non-Parametric Instance-level Discriminationcited by 1,845
- 2019FCOS: Fully Convolutional One-Stage Object Detectioncited by 1,577
- 2018Group Normalizationcited by 1,524
- 2017Accurate, Large Minibatch SGD: Training ImageNet in 1 Hourcited by 1,492
- 2016Context Encoders: Feature Learning by Inpaintingcited by 1,428
- 2016Unsupervised Learning of Visual Representations by Solving Jigsaw Puzzlescited by 1,382
- 2019Contrastive Multiview Codingcited by 1,301
- 2017Cascade R-CNN: Delving into High Quality Object Detectioncited by 1,276
- 2018Learning deep representations by mutual information estimation and maximizationcited by 1,273
- 2019LVIS: A Dataset for Large Vocabulary Instance Segmentationcited by 1,217
- 2016Instance Normalization: The Missing Ingredient for Fast Stylizationcited by 1,198
- 2016Colorful Image Colorizationcited by 1,114
- 2015Unsupervised Visual Representation Learning by Context Predictioncited by 1,066
- 2019Unsupervised Data Augmentation for Consistency Trainingcited by 1,060
- 2017Virtual Adversarial Training: A Regularization Method for Supervised and Semi-Supervised Learningcited by 920
- 2020What Makes for Good Views for Contrastive Learning?cited by 846
- 2019Self-Supervised Learning of Pretext-Invariant Representationscited by 808
- 2019Learning Representations by Maximizing Mutual Information Across Viewscited by 734
- 2019Data-Efficient Image Recognition with Contrastive Predictive Codingcited by 673
- 2020Prototypical Contrastive Learning of Unsupervised Representationscited by 640
- 2016Weight Normalization: A Simple Reparameterization to Accelerate Training of Deep Neural Networkscited by 543
- 2017Large Batch Training of Convolutional Networkscited by 535
- 2017Mean teachers are better role models: Weight-averaged consistency targets improve semi-supervised deep learning resultscited by 472
- 2019S4L: Self-Supervised Semi-Supervised Learningcited by 335
- 2014Discriminative Unsupervised Feature Learning with Exemplar Convolutional Neural Networkscited by 300
- 2019Large Scale Adversarial Representation Learningcited by 239
- 2016Split-Brain Autoencoders: Unsupervised Learning by Cross-Channel Predictioncited by 217
- 2019Local Aggregation for Unsupervised Learning of Visual Embeddingscited by 204
- 2017Multi-task Self-Supervised Visual Learningcited by 200
- 2016Learning Features by Watching Objects Movecited by 138
- 2017Batch Renormalization: Towards Reducing Minibatch Dependence in Batch-Normalized Modelscited by 128
- 2020BYOL works even without batch statisticscited by 95
- 2017MegDet: A Large Mini-Batch Object Detectorcited by 66
- 2019Online Normalization for Training Neural Networkscited by 29
- 2016Normalization Propagation: A Parametric Technique for Removing Internal Covariate Shift in Deep Networkscited by 20
50 most cited of 53 All 53 (JSON)
Built on by
Citing papers per year: 2021: 5; 2022: 7; 2023: 1; 2024: 1.
14 papers.
Sort:
By area:
- 2021Rethinking "Batch" in BatchNormcited by 45
- 2022On the Importance of Asymmetry for Siamese Representation Learningcited by 39
- 2023Multi-Mode Online Knowledge Distillation for Self-Supervised Visual Representation Learningcited by 20
- 2021Bag of Tricks and A Strong baseline for Image Copy Detectioncited by 14
- 2022On the Pros and Cons of Momentum Encoder in Self-Supervised Visual Representation Learningcited by 10
- 2021Temporal Knowledge Consistency for Unsupervised Visual Representation Learningcited by 8
- 2022Bootstrapped Representation Learning for Skeleton-Based Action Recognitioncited by 8
- 2022Semantics-Consistent Feature Search for Self-Supervised Visual Representation Learningcited by 8
- 2021D2LV: A Data-Driven and Local-Verification Approach for Image Copy Detectioncited by 7
- 2022Self-Supervised Visual Representation Learning via Residual Momentumcited by 7
- 2021Voxel-wise Cross-Volume Representation Learning for 3D Neuron Reconstructioncited by 4
- 2022Adversarial Contrastive Self-Supervised Learningcited by 2
2 more
- 2024MCA: Moment Channel Attention Networks4 Mar 2024
- 2023Multi-Mode Online Knowledge Distillation for Self-Supervised Visual Representation Learning13 Apr 2023
- 2022Semantics-Consistent Feature Search for Self-Supervised Visual Representation Learning13 Dec 2022
- 2022Self-Supervised Visual Representation Learning via Residual Momentum17 Nov 2022
- 2022On the Pros and Cons of Momentum Encoder in Self-Supervised Visual Representation Learning11 Aug 2022
- 2022DSPNet: Towards Slimmable Pretrained Networks based on Discriminative Self-supervised Learning13 Jul 2022
- 2022On the Importance of Asymmetry for Siamese Representation Learning1 Apr 2022
- 2022Adversarial Contrastive Self-Supervised Learning26 Feb 2022
- 2022Bootstrapped Representation Learning for Skeleton-Based Action Recognition4 Feb 2022
- 2021Bag of Tricks and A Strong baseline for Image Copy Detection13 Nov 2021
- 2021D2LV: A Data-Driven and Local-Verification Approach for Image Copy Detection13 Nov 2021
- 2021Temporal Knowledge Consistency for Unsupervised Visual Representation Learning24 Aug 2021
2 more
- 2021Voxel-wise Cross-Volume Representation Learning for 3D Neuron Reconstruction14 Aug 2021
- 2021Rethinking "Batch" in BatchNorm17 May 2021
- 2021Rethinking "Batch" in BatchNormcited by 45
- 2022On the Importance of Asymmetry for Siamese Representation Learningcited by 39
- 2021Bag of Tricks and A Strong baseline for Image Copy Detectioncited by 14
- 2022On the Pros and Cons of Momentum Encoder in Self-Supervised Visual Representation Learningcited by 10
- 2021Temporal Knowledge Consistency for Unsupervised Visual Representation Learningcited by 8
- 2022Bootstrapped Representation Learning for Skeleton-Based Action Recognitioncited by 8
- 2022Semantics-Consistent Feature Search for Self-Supervised Visual Representation Learningcited by 8
- 2021D2LV: A Data-Driven and Local-Verification Approach for Image Copy Detectioncited by 7
- 2022Self-Supervised Visual Representation Learning via Residual Momentumcited by 7
- 2021Voxel-wise Cross-Volume Representation Learning for 3D Neuron Reconstructioncited by 4
- 2022Adversarial Contrastive Self-Supervised Learningcited by 2
- 2024MCA: Moment Channel Attention Networkscited by 2
12 most cited of 13 All 13 (JSON)
- 2024MCA: Moment Channel Attention Networks4 Mar 2024
- 2022Semantics-Consistent Feature Search for Self-Supervised Visual Representation Learning13 Dec 2022
- 2022Self-Supervised Visual Representation Learning via Residual Momentum17 Nov 2022
- 2022On the Pros and Cons of Momentum Encoder in Self-Supervised Visual Representation Learning11 Aug 2022
- 2022DSPNet: Towards Slimmable Pretrained Networks based on Discriminative Self-supervised Learning13 Jul 2022
- 2022On the Importance of Asymmetry for Siamese Representation Learning1 Apr 2022
- 2022Adversarial Contrastive Self-Supervised Learning26 Feb 2022
- 2022Bootstrapped Representation Learning for Skeleton-Based Action Recognition4 Feb 2022
- 2021Bag of Tricks and A Strong baseline for Image Copy Detection13 Nov 2021
- 2021D2LV: A Data-Driven and Local-Verification Approach for Image Copy Detection13 Nov 2021
- 2021Temporal Knowledge Consistency for Unsupervised Visual Representation Learning24 Aug 2021
- 2021Voxel-wise Cross-Volume Representation Learning for 3D Neuron Reconstruction14 Aug 2021
12 newest of 13 All 13 (JSON)