HPCR: Holistic Proxy-based Contrastive Replay for Online Continual Learning

Online continual learning (OCL) aims to continuously learn new data from a single pass over the online data stream. It generally suffers from the catastrophic forgetting issue. Existing replay-based methods effectively alleviate this issue by replaying part of old data in a proxy-based or contrastive-based replay manner. In this paper, we conduct a comprehensive analysis of these two replay manners and find they can be complementary. Inspired by this finding, we propose a novel replay-based method called proxy-based contrastive replay (PCR), which replaces anchor-to-sample pairs with anchor-to-proxy pairs in the contrastive-based loss to alleviate the phenomenon of forgetting. Based on PCR, we further develop a more advanced method named holistic proxy-based contrastive replay (HPCR), which consists of three components. The contrastive component conditionally incorporates anchor-to-sample pairs to PCR, learning more fine-grained semantic information with a large training batch. The second is a temperature component that decouples the temperature coefficient into two parts based on their impacts on the gradient and sets different values for them to learn more novel knowledge. The third is a distillation component that constrains the learning process to keep more historical knowledge. Experiments on four datasets consistently demonstrate the superiority of HPCR over various state-of-the-art methods.

[1]  Xutao Li,et al.  UER: A Heuristic Bias Addressing Approach for Online Continual Learning , 2023, ACM Multimedia.

[2]  Hongming Shan,et al.  Online Prototype Learning for Online Continual Learning , 2023, 2023 IEEE/CVF International Conference on Computer Vision (ICCV).

[3]  Xutao Li,et al.  Anchor Assisted Experience Replay for Online Class-Incremental Learning , 2023, IEEE Transactions on Circuits and Systems for Video Technology.

[4]  Xutao Li,et al.  PCR: Proxy-Based Contrastive Replay for Online Class-Incremental Continual Learning , 2023, 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[5]  Jun Zhu,et al.  A Comprehensive Survey of Continual Learning: Theory, Method and Application , 2023, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[6]  B. Schiele,et al.  RMM: Reinforced Memory Management for Class-Incremental Learning , 2023, NeurIPS.

[7]  Jiwen Lu,et al.  Deep Metric Learning With Adaptively Composite Dynamic Constraints , 2023, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[8]  Yigang Cen,et al.  Contrastive Bayesian Analysis for Deep Metric Learning , 2022, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[9]  A. Bifet,et al.  A simple but strong baseline for online continual learning: Repeated Augmented Rehearsal , 2022, 2209.13917.

[10]  Qiuling Suo,et al.  Improving Task-free Continual Learning by Distributionally Robust Memory Evolution , 2022, ICML.

[11]  Dongyan Zhao,et al.  Adaptive Orthogonal Projection for Batch and Online Continual Learning , 2022, AAAI.

[12]  Ruiyu Li,et al.  PCL: Proxy-based Contrastive Learning for Domain Generalization , 2022, 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[13]  Kun-Juan Wei,et al.  Not Just Selection, but Exploration: Online Class-Incremental Continual Learning via Dual View Consistency , 2022, 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[14]  J. Choi,et al.  Hypergraph-Induced Semantic Tuplet Loss for Deep Metric Learning , 2022, 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[15]  Qixiang Ye,et al.  Dynamic Support Network for Few-Shot Class Incremental Learning , 2022, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[16]  L. Leal-Taixé,et al.  The Group Loss++: A Deeper Look Into Group Loss for Deep Metric Learning , 2022, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[17]  De-Chuan Zhan,et al.  Few-Shot Class-Incremental Learning by Sampling Multi-Phase Tasks , 2022, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[18]  S. Hoi,et al.  Continual Normalization: Rethinking Batch Normalization for Online Continual Learning , 2022, ICLR.

[19]  E. Ricci,et al.  Uncertainty-aware Contrastive Distillation for Incremental Semantic Segmentation , 2022, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[20]  Oriol Vinyals,et al.  Non-isotropy Regularization for Proxy-based Deep Metric Learning , 2022, 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[21]  Zhenguo Li,et al.  Memory Replay with Data Compression for Continual Learning , 2022, ICLR.

[22]  S. Calderara,et al.  Class-Incremental Continual Learning Into the eXtended DER-Verse , 2022, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[23]  Dacheng Tao,et al.  Hierarchical Prototype Networks for Continual Graph Representation Learning , 2021, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[24]  P. Indyk,et al.  Targeted Supervised Contrastive Learning for Long-Tailed Recognition , 2021, 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[25]  Jung-Woo Ha,et al.  Online Continual Learning on Class Incremental Blurry Task Configuration with Anytime Inference , 2021, ICLR.

[26]  Fengqing Zhu,et al.  Online Continual Learning For Visual Food Classification , 2021, 2021 IEEE/CVF International Conference on Computer Vision Workshops (ICCVW).

[27]  Jiwen Lu,et al.  Towards Interpretable Deep Metric Learning with Structural Matching , 2021, 2021 IEEE/CVF International Conference on Computer Vision (ICCV).

[28]  Jinwoo Shin,et al.  Co2L: Contrastive Continual Learning , 2021, 2021 IEEE/CVF International Conference on Computer Vision (ICCV).

[29]  Jungwon Lee,et al.  Dual-Teacher Class-Incremental Learning With Data-Free Generative Replay , 2021, 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW).

[30]  Guyue Hu,et al.  DeepCollaboration: Collaborative Generative and Discriminative Models for Class Incremental Learning , 2021, AAAI.

[31]  Dapeng Chen,et al.  Layerwise Optimization by Gradient Decomposition for Continual Learning , 2021, 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[32]  T. Tuytelaars,et al.  New Insights on Reducing Abrupt Representation Change in Online Continual Learning , 2021, International Conference on Learning Representations.

[33]  Jihwan Bang,et al.  Rainbow Memory: Continual Learning with a Memory of Diverse Samples , 2021, 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[34]  Scott Sanner,et al.  Supervised Contrastive Replay: Revisiting the Nearest Class Mean Classifier in Online Class-Incremental Continual Learning , 2021, 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW).

[35]  Xiaokang Yang,et al.  Continual Learning for Blind Image Quality Assessment , 2021, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[36]  Hyunwoo J. Kim,et al.  Online Continual Learning in Image Classification: An Empirical Survey , 2021, Neurocomputing.

[37]  Feng Wang,et al.  Understanding the Behaviour of Contrastive Loss , 2020, 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[38]  Adrian Popescu,et al.  A Comprehensive Study of Class Incremental Learning Algorithms for Visual Tasks , 2020, Neural Networks.

[39]  Joost van de Weijer,et al.  Class-Incremental Learning: Survey and Performance Evaluation on Image Classification , 2020, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[40]  Simone Calderara,et al.  Rethinking Experience Replay: a Bag of Tricks for Continual Learning , 2020, 2020 25th International Conference on Pattern Recognition (ICPR).

[41]  Seyed Iman Mirzadeh,et al.  Linear Mode Connectivity in Multitask and Continual Learning , 2020, ICLR.

[42]  Scott Sanner,et al.  Online Class-Incremental Continual Learning with Adversarial Shapley Value , 2020, AAAI.

[43]  Philip H. S. Torr,et al.  GDumb: A Simple Approach that Questions Our Progress in Continual Learning , 2020, ECCV.

[44]  Hava T. Siegelmann,et al.  Brain-inspired replay for continual learning with artificial neural networks , 2020, Nature Communications.

[45]  V. Koltun,et al.  Drinking From a Firehose: Continual Learning With Web-Scale Natural Language , 2020, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[46]  Xiang Ren,et al.  Gradient-based Editing of Memory Examples for Online Task-free Continual Learning , 2020, NeurIPS.

[47]  S. Calderara,et al.  Dark Experience for General Continual Learning: a Strong, Simple Baseline , 2020, NeurIPS.

[48]  Björn Ommer,et al.  Sharing Matters for Generalization in Deep Metric Learning , 2020, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[49]  Taesup Moon,et al.  SS-IL: Separated Softmax for Incremental Learning , 2020, 2021 IEEE/CVF International Conference on Computer Vision (ICCV).

[50]  Shutao Xia,et al.  Maintaining Discrimination and Fairness in Class Incremental Learning , 2019, 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[51]  Mehrdad Farajtabar,et al.  Orthogonal Gradient Descent for Continual Learning , 2019, AISTATS.

[52]  Ying Fu,et al.  Incremental Learning Using Conditional Adversarial Networks , 2019, 2019 IEEE/CVF International Conference on Computer Vision (ICCV).

[53]  Albert Gordo,et al.  Using Hindsight to Anchor Past Knowledge in Continual Learning , 2019, AAAI.

[54]  Matthias De Lange,et al.  A Continual Learning Survey: Defying Forgetting in Classification Tasks , 2019, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[55]  Tinne Tuytelaars,et al.  Online Continual Learning with Maximally Interfered Retrieval , 2019, ArXiv.

[56]  Dahua Lin,et al.  Learning a Unified Classifier Incrementally via Rebalancing , 2019, 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[57]  Yoshua Bengio,et al.  Gradient based sample selection for online continual learning , 2019, NeurIPS.

[58]  Tinne Tuytelaars,et al.  Task-Free Continual Learning , 2018, 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[59]  David Rolnick,et al.  Experience Replay for Continual Learning , 2018, NeurIPS.

[60]  Rama Chellappa,et al.  Learning Without Memorizing , 2018, 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[61]  Marc'Aurelio Ranzato,et al.  Efficient Lifelong Learning with A-GEM , 2018, ICLR.

[62]  Marc'Aurelio Ranzato,et al.  Gradient Episodic Memory for Continual Learning , 2017, NIPS.

[63]  Andrei A. Rusu,et al.  Overcoming catastrophic forgetting in neural networks , 2016, Proceedings of the National Academy of Sciences.

[64]  Christoph H. Lampert,et al.  iCaRL: Incremental Classifier and Representation Learning , 2016, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[65]  Derek Hoiem,et al.  Learning without Forgetting , 2016, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[66]  Razvan Pascanu,et al.  Progressive Neural Networks , 2016, ArXiv.

[67]  Oriol Vinyals,et al.  Matching Networks for One Shot Learning , 2016, NIPS.

[68]  Geoffrey E. Hinton,et al.  Distilling the Knowledge in a Neural Network , 2015, ArXiv.

[69]  Gabriela Csurka,et al.  Distance-Based Image Classification: Generalizing to New Classes at Near-Zero Cost , 2013, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[70]  Philip H. S. Torr,et al.  Real-Time Evaluation in Online Continual Learning: A New Paradigm , 2023, ArXiv.

[71]  A. Chrysakis,et al.  Online Bias Correction for Task-Free Continual Learning , 2023, ICLR.

[72]  B. Liu,et al.  Online Continual Learning through Mutual Information Maximization , 2022, ICML.

[73]  Qiang Qiu,et al.  Continual Learning with Filter Atom Swapping , 2022, ICLR.

[74]  Peng Yang,et al.  Mitigating Forgetting in Online Continual Learning with Neuron Calibration , 2021, NeurIPS.

[75]  Xu-Yao Zhang,et al.  Class-Incremental Learning via Dual Augmentation , 2021, NeurIPS.

[76]  Ya Le,et al.  Tiny ImageNet Visual Recognition Challenge , 2015 .

[77]  Alex Krizhevsky,et al.  Learning Multiple Layers of Features from Tiny Images , 2009 .

[78]  Geoffrey E. Hinton,et al.  Visualizing Data using t-SNE , 2008 .