Article

Online Continual Learning Under Extreme Memory Constraints

Authors:

Stéphane Lathuilière,

Enver Sangineto,

Elisa RicciAuthors Info & Claims

Computer Vision – ECCV 2020: 16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part XXVIII

Pages 720 - 735

https://doi.org/10.1007/978-3-030-58604-1_43

Published: 23 August 2020 Publication History

Abstract

Continual Learning (CL) aims to develop agents emulating the human ability to sequentially learn new tasks while being able to retain knowledge obtained from past experiences. In this paper, we introduce the novel problem of Memory-Constrained Online Continual Learning (MC-OCL) which imposes strict constraints on the memory overhead that a possible algorithm can use to avoid catastrophic forgetting. As most, if not all, previous CL methods violate these constraints, we propose an algorithmic solution to MC-OCL: Batch-level Distillation (BLD), a regularization-based CL approach, which effectively balances stability and plasticity in order to learn from data streams, while preserving the ability to solve old tasks through distillation. Our extensive experimental evaluation, conducted on three publicly available benchmarks, empirically demonstrates that our approach successfully addresses the MC-OCL problem and achieves comparable accuracy to prior distillation methods requiring higher memory overhead (Code available at https://github.com/DonkeyShot21/batch-level-distillation).

References

[1]

Aljundi R, Babiloni F, Elhoseiny M, Rohrbach M, and Tuytelaars T Ferrari V, Hebert M, Sminchisescu C, and Weiss Y Memory aware synapses: learning what (not) to forget Computer Vision – ECCV 2018 2018 Cham Springer 144-161

[2]

Aljundi, R., et al.: Online continual learning with maximal interfered retrieval. In: NeurIPS, pp. 11849–11860 (2019)

[3]

Aljundi, R., Kelchtermans, K., Tuytelaars, T.: Task-free continual learning. In: CVPR (2019)

[4]

Aljundi, R., Lin, M., Goujaud, B., Bengio, Y.: Gradient based sample selection for online continual learning. In: NeurIPS (2019)

[5]

Castro FM, Marín-Jiménez MJ, Guil N, Schmid C, and Alahari K Ferrari V, Hebert M, Sminchisescu C, and Weiss Y End-to-end incremental learning Computer Vision – ECCV 2018 2018 Cham Springer 241-257

[6]

Cermelli, F., Mancini, M., Bulo, S.R., Ricci, E., Caputo, B.: Modeling the background for incremental learning in semantic segmentation. In: CVPR, pp. 9233–9242 (2020)

[7]

Chaudhry A, Dokania PK, Ajanthan T, and Torr PHS Ferrari V, Hebert M, Sminchisescu C, and Weiss Y Riemannian walk for incremental learning: understanding forgetting and intransigence Computer Vision – ECCV 2018 2018 Cham Springer 556-572

[8]

Dhar, P., Singh, R.V., Peng, K.C., Wu, Z., Chellappa, R.: Learning without memorizing. In: CVPR (2019)

[9]

Duda RO, Hart PE, and Stork DG Pattern Classification 2000 2 New York Wiley

[10]

Farquhar, S., Gal, Y.: Towards robust evaluations of continual learning. arXiv preprint arXiv:1805.09733 (2018)

[11]

He, K., Zhang, X., Ren, S., Sun, J.: Deep residual learning for image recognition. In: CVPR, pp. 770–778 (2016)

[12]

Hinton, G., Vinyals, O., Dean, J.: Distilling the knowledge in a neural network. stat (2015)

[13]

Hou, S., Pan, X., Loy, C.C., Wang, Z., Lin, D.: Learning a unified classifier incrementally via rebalancing. In: CVPR (2019)

[14]

Javed, K., White, M.: Meta-learning representations for continual learning. In: NeurIPS, pp. 1820–1830 (2019)

[15]

Kirkpatrick, J., et al.: Overcoming catastrophic forgetting in neural networks. In: PNAS (2017)

[16]

Krizhevsky, A.: Learning multiple layers of features from tiny images. Technical report (2009)

[17]

Lange, M.D., et al.: Continual learning: a comparative study on how to defy forgetting in classification tasks. arXiv:1909.08383 (2019)

[18]

LeCun, Y., Cortes, C.: MNIST handwritten digit database (2010). http://yann.lecun.com/exdb/mnist/

[19]

Lee, S., Ha, J., Zhang, D., Kim, G.: A neural dirichlet process mixture model for task-free continual learning. In: ICLR (2020)

[20]

Li Z and Hoiem D Learning without forgetting IEEE T-PAMI 2017 40 12 2935-2947

[21]

Lopez-Paz, D., Ranzato, M.: Gradient episodic memory for continual learning. In: NIPS (2017)

[22]

Mallya, A., Lazebnik, S.: PackNet: adding multiple tasks to a single network by iterative pruning. In: CVPR (2018)

[23]

Netzer, Y., Wang, T., Coates, A., Bissacco, A., Wu, B., Ng, A.Y.: Reading digits in natural images with unsupervised feature learning. In: NIPS (2011)

[24]

Ostapenko, O., Puscas, M., Klein, T., Jahnichen, P., Nabi, M.: Learning to remember: a synaptic plasticity driven framework for continual learning. In: CVPR (2019)

[25]

Paszke, A., et al.: Automatic differentiation in pytorch (2017)

[26]

Rebuffi, S.A., Kolesnikov, A., Sperl, G., Lampert, C.H.: iCaRL: incremental classifier and representation learning. In: CVPR (2017)

[27]

Riemer, M., et al.: Learning to learn without forgetting by maximizing transfer and minimizing interference. arXiv preprint arXiv:1810.11910 (2018)

[28]

Rusu, A.A., et al.: Progressive neural networks. arXiv preprint arXiv:1606.04671 (2016)

[29]

Shin, H., Lee, J.K., Kim, J., Kim, J.: Continual learning with deep generative replay. In: NeurIPS (2017)

[30]

Wu, C., Herranz, L., Liu, X., van de Weijer, J., Raducanu, B., et al.: Memory replay GANs: learning to generate new categories without forgetting. In: NeurIPS (2018)

[31]

Wu, Y., et al.: Large scale incremental learning. In: CVPR (2019)

Cited By

Li QPeng YZhou JCai JKankanhalli MPrabhakaran BBoll SSubramanian RZheng LSingh VCesar PXie LXu D(2024)Progressive Prototype Evolving for Dual-Forgetting Mitigation in Non-Exemplar Online Continual LearningProceedings of the 32nd ACM International Conference on Multimedia10.1145/3664647.3681234(2477-2486)Online publication date: 28-Oct-2024
https://dl.acm.org/doi/10.1145/3664647.3681234
Li YXu WWang HQi YGuo JLi R(2024)Personalized Federated Domain-Incremental Learning Based on Adaptive Knowledge MatchingComputer Vision – ECCV 202410.1007/978-3-031-72952-2_8(127-144)Online publication date: 29-Sep-2024
https://dl.acm.org/doi/10.1007/978-3-031-72952-2_8
Neff CDanesh Pazho ATabkhi H(2023)Real-time online unsupervised domain adaptation for real-world person re-identificationJournal of Real-Time Image Processing10.1007/s11554-023-01362-z20:6Online publication date: 24-Sep-2023
https://dl.acm.org/doi/10.1007/s11554-023-01362-z
Show More Cited By

Index Terms

Online Continual Learning Under Extreme Memory Constraints
1. Computing methodologies
  1. Artificial intelligence
  2. Machine learning
2. Theory of computation
  1. Theory and algorithms for application domains

Index terms have been assigned to the content through auto-classification.

Recommendations

Avalanche RL: A Continual Reinforcement Learning Library
Image Analysis and Processing – ICIAP 2022
Abstract
Continual Reinforcement Learning (CRL) is a challenging setting where an agent learns to interact with an environment that is constantly changing over time (the stream of experiences). In this paper, we describe Avalanche RL, a library for ...
Continual learning for robotics: Definition, framework, learning strategies, opportunities and challenges
Highlights
- State of the art on continual / lifelong learning and its implications for robotics.
- Proposal of a framework to present Continual Learning algorithms.
- Summary of existing metrics for continual learning.
- Presenting robotics ...
Abstract
Continual learning (CL) is a particular machine learning paradigm where the data distribution and learning objective change through time, or where all the training data and objective criteria are never available at once. The evolution of the ...
Towards continual reinforcement learning through evolutionary meta-learning
GECCO '19: Proceedings of the Genetic and Evolutionary Computation Conference Companion

In continual learning, an agent is exposed to a changing environment, requiring it to adapt during execution time. While traditional reinforcement learning (RL) methods have shown impressive results in various domains, there has been less progress in ...

Comments

Information & Contributors

Information

Published In

cover image Guide Proceedings

Computer Vision – ECCV 2020: 16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part XXVIII

Aug 2020

829 pages

ISBN:978-3-030-58603-4

DOI:10.1007/978-3-030-58604-1

Editors:
Andrea Vedaldi
University of Oxford, Oxford, UK
,
Horst Bischof
Graz University of Technology, Graz, Austria
,
Thomas Brox
University of Freiburg, Freiburg im Breisgau, Germany
,
Jan-Michael Frahm
University of North Carolina at Chapel Hill, Chapel Hill, NC, USA

© Springer Nature Switzerland AG 2020.

Publisher

Springer-Verlag

Berlin, Heidelberg

Publication History

Published: 23 August 2020

Author Tags

Qualifiers

Article

Contributors

Other Metrics

View Article Metrics

Bibliometrics & Citations

Bibliometrics

Article Metrics

5
Total Citations
View Citations
0
Total Downloads

Downloads (Last 12 months)0
Downloads (Last 6 weeks)0

Reflects downloads up to 08 Mar 2025

Other Metrics

View Author Metrics

Citations

Cited By

Li QPeng YZhou JCai JKankanhalli MPrabhakaran BBoll SSubramanian RZheng LSingh VCesar PXie LXu D(2024)Progressive Prototype Evolving for Dual-Forgetting Mitigation in Non-Exemplar Online Continual LearningProceedings of the 32nd ACM International Conference on Multimedia10.1145/3664647.3681234(2477-2486)Online publication date: 28-Oct-2024
https://dl.acm.org/doi/10.1145/3664647.3681234
Li YXu WWang HQi YGuo JLi R(2024)Personalized Federated Domain-Incremental Learning Based on Adaptive Knowledge MatchingComputer Vision – ECCV 202410.1007/978-3-031-72952-2_8(127-144)Online publication date: 29-Sep-2024
https://dl.acm.org/doi/10.1007/978-3-031-72952-2_8
Neff CDanesh Pazho ATabkhi H(2023)Real-time online unsupervised domain adaptation for real-world person re-identificationJournal of Real-Time Image Processing10.1007/s11554-023-01362-z20:6Online publication date: 24-Sep-2023
https://dl.acm.org/doi/10.1007/s11554-023-01362-z
Wang ZLi DLi PAl Hasan MXiong L(2022)Latent Coreset Sampling based Data-Free Continual LearningProceedings of the 31st ACM International Conference on Information & Knowledge Management10.1145/3511808.3557375(2077-2087)Online publication date: 17-Oct-2022
https://dl.acm.org/doi/10.1145/3511808.3557375
Jiang MLi F(2022)Lie group continual meta learning algorithmApplied Intelligence10.1007/s10489-021-03036-452:10(10965-10978)Online publication date: 18-Jan-2022
https://dl.acm.org/doi/10.1007/s10489-021-03036-4

View Options

View options

Figures

Tables

Media

View Table of Conten