Abstract
Deep neural networks have made outstanding achievements in many static tasks, however, when faced with incremental scenario, they suffer from catastrophic forgetting since the previous data is usually inaccessible. Stored data and generative models are commonly used for maintaining model performance, but there exist problems of memory utilization and privacy safety. In this paper, a novel non-exemplar based incremental learning model, Prototype Representation Expansion (PRE), which provides a great degree to retain the feature space of old tasks, is proposed. Firstly, prototypes are generated to meet the stability and robustness. The mean value of feature embedding for each class is used as prototype to maintain the model stability. Meanwhile, PRE also selects prototypes according to their responses of classifier by feature disturbance noise injection, and the decision boundary can better be maintained. Secondly, prototypes of various classes are linearly combined to construct the hybrid prototype with mixed labels. Along with prototype augmentation, they are used for incremental training phase. We conduct extensive experiments on two benchmark datasets, CIFAR-100 and ImageNet-Subset. It shows that PRE can be combined with some non-exemplar based methods to significantly improve their ability and achieve comparable performance to exemplar based methods.
Similar content being viewed by others
Data Availibility
Data openly available in a public repository. The data that support the findings of this study are openly available at: https://image-net.org/ and https://www.cs.toronto.edu/~kriz/cifar.html
References
Aljundi R, Babiloni F, Elhoseiny M, Rohrbach M, Tuytelaars T (2018) Memory aware synapses: Learning what (not) to forget. ECCV, Lect Notes Comput Sci 11207:144–161. https://doi.org/10.1007/978-3-030-01219-9_9
Aljundi, R, Chakravarty P, Tuytelaars T (2017) Expert gate: lifelong learning with a network of experts. In: IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp 7120–7129. IEEE Computer Society . https://doi.org/10.1109/CVPR.2017.753
Bang J, Kim H, Yoo Y, Ha JW, Choi J (2021) Rainbow memory: continual learning with a memory of diverse samples. In: IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp 8218–8227
Berthelot D, Carlini N, Goodfellow IJ, Papernot N, Oliver A, Raffel C (2019) Mixmatch: a holistic approach to semi-supervised learning. In: Advances in Neural Information Processing Systems, pp 5050–5060 .https://proceedings.neurips.cc/paper/2019/hash/1cd138d0499a68f4bb72bee04bbec2d7-Abstract.html
Castro FM, Marín-Jiménez MJ, Guil N, Schmid C, Alahari K (2018) End-to-end incremental learning. ECCV 11216:241–257. https://doi.org/10.1007/978-3-030-01258-8_15
Chaudhry A, Dokania PK, Ajanthan T, Torr PHS (2018) Riemannian walk for incremental learning: Understanding forgetting and intransigence. In: V. Ferrari, M. Hebert, C. Sminchisescu, Y. Weiss (eds.) ECCV, Lecture Notes in Computer Science, vol. 11215, pp. 556–572. Springer . https://doi.org/10.1007/978-3-030-01252-6_33
He K, Zhang X, Ren S, Sun J (2016) Deep residual learning for image recognition. In: IEEE Conference on Computer Vision and Pattern Recognition, pp 770–778
Hinton GE, Vinyals O, Dean J (2015) Distilling the knowledge in a neural network. ArXiv abs/1503.02531
Hou S, Pan X, Loy CC, Wang Z, Lin D (2018) Lifelong learning via progressive distillation and retrospection. In: Proceedings of the European Conference on Computer Vision (ECCV)
Hou S, Pan X, Loy CC, Wang Z, Lin D (2019) Learning a unified classifier incrementally via rebalancing. In: IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp 831–839
Kemker R, McClure M, Abitino A, Hayes TL, Kanan C (2018) Measuring catastrophic forgetting in neural networks. In: Proceedings of the Thirty-Second AAAI Conference on Artificial Intelligence, pp 3390–3398. AAAI Press . https://www.aaai.org/ocs/index.php/AAAI/AAAI18/paper/view/16410
Kingma DP, Ba J (2015) Adam: a method for stochastic optimization. In: Y. Bengio, Y. LeCun (eds.) 3rd International Conference on Learning Representations, ICLR 2015, San Diego, CA, USA, May 7-9, 2015, Conference Track Proceedings. http://arxiv.org/abs/1412.6980
Kirkpatrick J, Pascanu R, Rabinowitz NC, Veness J, Desjardins G, Rusu AA, Milan K, Quan J, Ramalho T, Grabska-Barwinska A, Hassabis D, Clopath C, Kumaran D, Hadsell R (2017) Overcoming catastrophic forgetting in neural networks. Proc Natl Acad Sci 114:3521–3526
Krizhevsky A (2009) Learning multiple layers of features from tiny images
Lee K, Lee K, Shin J, Lee H (2019) Overcoming catastrophic forgetting with unlabeled data in the wild. In: IEEE/CVF International Conference on Computer Vision (ICCV), pp 312–321. IEEE . https://doi.org/10.1109/ICCV.2019.00040
Li Z, Hoiem D (2018) Learning without forgetting. IEEE Trans Pattern Anal Mach Intell 40:2935–2947
Liu Y, Liu A, Su Y, Schiele B, Sun Q (2020) Mnemonics training: multi-class incremental learning without forgetting. In: IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp 12242–12251
Liu Y, Parisot S, Slabaugh GG, Jia X, Leonardis A, Tuytelaars T ()2020 More classifiers, less forgetting: a generic multi-classifier paradigm for incremental learning. In: A. Vedaldi, H. Bischof, T. Brox, J. Frahm (eds.) ECCV, Lecture Notes in Computer Science, vol. 12371, pp. 699–716. Springer . https://doi.org/10.1007/978-3-030-58574-7_42
Lomonaco V, Maltoni D (2017) Core50: a new dataset and benchmark for continuous object recognition. In: CoRL, Proceedings of Machine Learning Research, vol. 78, pp. 17–26. PMLR. http://proceedings.mlr.press/v78/lomonaco17a.html
McCloskey M, Cohen N (1989) Catastrophic interference in connectionist networks: the sequential learning problem. Psychol Learn Motiv- Adv Res Theory 24(C):109–165. https://doi.org/10.1016/S0079-7421(08)60536-8
McCloskey M, Cohen N (1989) Catastrophic interference in connectionist networks: the sequential learning problem. Psychol Learn Motiv 24:109–165
Mnih V, Kavukcuoglu K, Silver D, Rusu AA, Veness J, Bellemare MG, Graves A, Riedmiller MA, Fidjeland A, Ostrovski G, Petersen S, Beattie C, Sadik A, Antonoglou I, King H, Kumaran D, Wierstra D, Legg S, Hassabis D (2015) Human-level control through deep reinforcement learning. Nature 518(7540):529–533. https://doi.org/10.1038/nature14236
Park DS, Chan W, Zhang Y, Chiu C, Zoph B, Cubuk ED, Le QV (2019) Specaugment: a simple data augmentation method for automatic speech recognition. In: Interspeech 2019, 20th Annual Conference of the International Speech Communication Association, Graz, Austria, 15-19 September 2019, pp 2613–2617. ISCA . https://doi.org/10.21437/Interspeech.2019-2680
Ratcliff R (1990) Connectionist models of recognition memory: constraints imposed by learning and forgetting functions. Psychol Rev 97(2):285–308
Rebuffi SA, Kolesnikov A, Sperl G, Lampert CH (2017) ICARL: incremental classifier and representation learning. In: IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp 5533–5542
Robins AV (1993) Catastrophic forgetting in neural networks: the role of rehearsal mechanisms. In: Proceedings 1993 The First New Zealand International Two-Stream Conference on Artificial Neural Networks and Expert Systems pp 65–68
Robins AV (1995) Catastrophic forgetting, rehearsal and pseudorehearsal. Connect Sci 7:123–146
Roy D, Panda P, Roy K (2020) Tree-CNN: a hierarchical deep convolutional neural network for incremental learning. Neural Netw 121:148–160. https://doi.org/10.1016/j.neunet.2019.09.010
Russakovsky O, Deng J, Su H, Krause J, Satheesh S, Ma S, Huang Z, Karpathy A, Khosla A, Bernstein MS, Berg A, Fei-Fei L (2015) Imagenet large scale visual recognition challenge. Int J Comput Vision 115:211–252
Rusu AA, Rabinowitz NC, Desjardins G, Soyer H, Kirkpatrick J, Kavukcuoglu K, Pascanu R, Hadsell R (2016) Progressive neural networks. ArXiv abs/1606.04671
Silver D, Huang A, Maddison CJ, Guez A, Sifre L, van den Driessche G, Schrittwieser J, Antonoglou I, Panneershelvam V, Lanctot M, Dieleman S, Grewe D, Nham J, Kalchbrenner N, Sutskever I, Lillicrap TP, Leach M, Kavukcuoglu K, Graepel T, Hassabis D (2016) Mastering the game of go with deep neural networks and tree search. Nature 529(7587):484–489. https://doi.org/10.1038/nature16961
van de Ven GM, Tolias A (2019) Three scenarios for continual learning. ArXiv abs/1904.07734
Wei JW, Zou K (2019) Eda: Easy data augmentation techniques for boosting performance on text classification tasks. In: EMNLP-IJCNLP, pp 6381–6387. Association for Computational Linguistics . https://doi.org/10.18653/v1/D19-1670
Wu Y, Chen Y, Wang L, Ye Y, Liu Z, Guo Y, Fu Y (2019) Large scale incremental learning. In: IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp 374–382. Computer Vision Foundation / IEEE . https://doi.org/10.1109/CVPR.2019.00046. http://openaccess.thecvf.com/content_CVPR_2019/html/Wu_Large_Scale_Incremental_Learning_CVPR_2019_paper.html
Yan S, Xie J, He X (2021) Der: Dynamically expandable representation for class incremental learning. In: IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp 3014–3023.
Yoon J, Yang E, Lee J, Hwang SJ (2018) Lifelong learning with dynamically expandable networks. In: International Conference on Learning Representationss. OpenReview.net . https://openreview.net/forum?id=Sk7KsfW0-
Yu L, Twardowski B, Liu X, Herranz L, Wang K, mei Cheng Y, Jui S, van de Weijer J (2020) Semantic drift compensation for class-incremental learning. In: IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp 6980–6989
Zeng G, Chen Y, Cui B, Yu S (2019) Continual learning of context-dependent processing in neural networks. Nat Mach Intell 1(8):364–372. https://doi.org/10.1038/s42256-019-0080-x
Zenke F, Poole B, Ganguli S (2017) Continual learning through synaptic intelligence. Proc Mach Learn Res 70:3987–3995
Zhang H, Cissé M, Dauphin YN, Lopez-Paz D (2018) MIXUP: beyond empirical risk minimization. In: International Conference on Learning Representations. OpenReview.net . https://openreview.net/forum?id=r1Ddp1-Rb
Zhao B, Xiao X, Gan G, Zhang B, Xia S (2020) Maintaining disclrimination and fairness in class incremental learning. In: IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp 13205–13214
Zhu F, Zhang X, Wang C, Yin F, Liu C (2021) Prototype augmentation and self-supervision for incremental learning. In: IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp 5871–5880. Computer Vision Foundation / IEEE. https://openaccess.thecvf.com/content/CVPR2021/html/Zhu_Prototype_Augmentation_and_Self-Supervision_for_Incremental_Learning_CVPR_2021_paper.html
Funding
This research is supported by Natural Science Foundation of Liaoning Province, China (No.2022-MS-112).
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflicts of interest
The authors declare that they have no conflict of interest.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Mao, K., Luo, Y., Ren, Y. et al. Prototype Representation Expansion in Incremental Learning. Neural Process Lett 55, 8401–8417 (2023). https://doi.org/10.1007/s11063-023-11317-x
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11063-023-11317-x