Conclusion
We have presented PyCIL, a classincremental learning toolbox written in Python. It contains implementations of a number of founding studies of CIL, but also provides current state-of-the-art algorithms that can be used to conduct novel fundamental research. Code consistency makes it an easy tool for research purposes, teaching, and industrial applications.
Article PDF
References
Kirkpatrick J, Pascanu R, Rabinowitz N, et al. Overcoming catastrophic forgetting in neural networks. Proc Natl Acad Sci USA, 2017, 114: 3521–3526
Li Z Z, Hoiem D. Learning without forgetting. IEEE Trans Pattern Anal Mach Intell, 2018, 40: 2935–2947
Rebuffi S A, Kolesnikov A, Sperl G, et al. iCaRL: incremental classifier and representation learning. In: Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, 2017. 2001–2010
Lopez-Paz D, Ranzato M. Gradient episodic memory for continual learning. In: Proceedings of International Conference on Neural Information Processing Systems, 2017. 6467–6476
Wu Y, Chen Y, Wang L, et al. Large scale incremental learning. In: Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, 2019. 374–382
Zhao B, Xiao X, Gan G, et al. Maintaining discrimination and fairness in class incremental learning. In: Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, 2020. 13208–13217
Douillard A, Cord M, Ollion C, et al. PODNet: pooled outputs distillation for small-tasks incremental learning. In: Proceedings of European Conference on Computer Vision, 2020. 86–102
Yan S, Xie J, He X. DER: dynamically expandable representation for class incremental learning. In: Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, 2021. 3014–3023
Zhou D W, Ye H J, Zhan D C. Co-transport for class-incremental learning. In: Proceedings of ACM International Conference on Multimedia, 2021. 1645–1654
Acknowledgements
This work was supported by National Key R&D Program of China (Grant No. 2020AAA0109401), National Natural Science Foundation of China (Grant Nos. 61773198, 61921006, 62006112), NSFC-NRF Joint Research Project (Grant No. 61861146001), Collaborative Innovation Center of Novel Software Technology and Industrialization, and Natural Science Foundation of Jiangsu Province (Grant No. BK20200313).
Author information
Authors and Affiliations
Corresponding author
Additional information
Zhou D-W and Wang F-Y have the same contribution to this work.
Rights and permissions
About this article
Cite this article
Zhou, DW., Wang, FY., Ye, HJ. et al. PyCIL: a Python toolbox for class-incremental learning. Sci. China Inf. Sci. 66, 197101 (2023). https://doi.org/10.1007/s11432-022-3600-y
Received:
Accepted:
Published:
DOI: https://doi.org/10.1007/s11432-022-3600-y