ABSTRACT
In Deep Neural Network training, the availability of a large amount of representative training data is the sine qua non-condition for a good generalization capacity of the model. In many real-world applications, data is not available at a glance, but coming on the fly. If a pre-trained model is fine-tuned on the new data, then catastrophic forgetting happens mostly. Incremental learning mechanisms propose ways to overcome catastrophic forgetting. Streaming learning is a type of incremental learning where models learn from new data instances as soon as they become available in a single training pass. In this work, we conduct an experimental study, on a large dataset, of an incremental/streaming learning method Move-to-Data we previously proposed, and propose an updated approach by ”re-targeting” with gradient descent which is faster than the popular streaming learning method ExStream. The method achieves better performances and computational efficiency compared to ExStream. Move-to-Data with gradient is on average 3.5 times faster than ExStream and has a similar accuracy, with 0.5% improvement compared to ExStream.
- Francisco M Castro, Manuel J Marín-Jiménez, Nicolás Guil, Cordelia Schmid, and Karteek Alahari. 2018. End-to-end incremental learning. In Proceedings of the European conference on computer vision (ECCV). 233–248.Google ScholarDigital Library
- Francisco M. Castro, Manuel J. Marín-Jiménez, Nicolás Guil, Cordelia Schmid, and Karteek Alahari. 2018. End-to-End Incremental Learning. In Computer Vision - ECCV 2018 - 15th European Conference, Munich, Germany, September 8-14, 2018, Proceedings, Part XII(Lecture Notes in Computer Science, Vol. 11216), Vittorio Ferrari, Martial Hebert, Cristian Sminchisescu, and Yair Weiss (Eds.). Springer, 241–257. https://doi.org/10.1007/978-3-030-01258-8_15Google ScholarDigital Library
- Anuvabh Dutt, Denis Pellerin, and Georges Quénot. 2017. Improving Hierarchical Image Classification with Merged CNN Architectures. In CBMI. ACM, 31:1–31:7.Google Scholar
- Robert M French. 1999. Catastrophic forgetting in connectionist networks. Trends in cognitive sciences 3, 4 (1999), 128–135.Google Scholar
- Tommaso Furlanello, Jiaping Zhao, Andrew M Saxe, Laurent Itti, and Bosco S Tjan. 2016. Active long term memory networks. arXiv preprint arXiv:1606.02355(2016).Google Scholar
- Tyler L Hayes, Nathan D Cahill, and Christopher Kanan. 2019. Memory efficient experience replay for streaming learning. In 2019 International Conference on Robotics and Automation (ICRA). IEEE, 9769–9776.Google ScholarDigital Library
- Tyler L Hayes, Kushal Kafle, Robik Shrestha, Manoj Acharya, and Christopher Kanan. 2020. Remind your neural network to prevent catastrophic forgetting. In European Conference on Computer Vision. Springer, 466–483.Google ScholarDigital Library
- Tyler L Hayes and Christopher Kanan. 2020. Lifelong machine learning with deep streaming linear discriminant analysis. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition workshops. 220–221.Google ScholarCross Ref
- Geoffrey Hinton, Oriol Vinyals, Jeff Dean, 2015. Distilling the knowledge in a neural network. arXiv preprint arXiv:1503.02531 2, 7 (2015).Google Scholar
- Heechul Jung, Jeongwoo Ju, Minju Jung, and Junmo Kim. 2016. Less-forgetting learning in deep neural networks. arXiv preprint arXiv:1607.00122(2016).Google Scholar
- Alex Krizhevsky, Geoffrey Hinton, 2009. Learning multiple layers of features from tiny images. (2009).Google Scholar
- Ya Le and Xuan Yang. 2015. Tiny imagenet visual recognition challenge. CS 231N 7, 7 (2015), 3.Google Scholar
- Kimin Lee, Kibok Lee, Honglak Lee, and Jinwoo Shin. 2018. A simple unified framework for detecting out-of-distribution samples and adversarial attacks. Advances in neural information processing systems 31 (2018).Google Scholar
- Zhizhong Li and Derek Hoiem. 2017. Learning without forgetting. IEEE transactions on pattern analysis and machine intelligence 40, 12(2017), 2935–2947.Google Scholar
- Edwin Lughofer. 2008. Extensions of vector quantization for incremental clustering. Pattern recognition 41, 3 (2008), 995–1011.Google Scholar
- Miltiadis Poursanidis, Jenny Benois-Pineau, Akka Zemmari, Boris Mansenca, and Aymar de Rugy. 2020. Move-to-Data: A new Continual Learning approach with Deep CNNs, Application for image-class recognition. arXiv preprint arXiv:2006.07152(2020).Google Scholar
- Amal Rannen, Rahaf Aljundi, Matthew B Blaschko, and Tinne Tuytelaars. 2017. Encoder based lifelong learning. In Proceedings of the IEEE International Conference on Computer Vision. 1320–1328.Google ScholarCross Ref
- Sylvestre-Alvise Rebuffi, Alexander Kolesnikov, Georg Sperl, and Christoph H Lampert. 2017. icarl: Incremental classifier and representation learning. In Proceedings of the IEEE conference on Computer Vision and Pattern Recognition. 2001–2010.Google ScholarCross Ref
- Olga Russakovsky, Jia Deng, Hao Su, Jonathan Krause, Sanjeev Satheesh, Sean Ma, Zhiheng Huang, Andrej Karpathy, Aditya Khosla, Michael Bernstein, 2015. Imagenet large scale visual recognition challenge. International journal of computer vision 115, 3 (2015), 211–252.Google ScholarDigital Library
- Jason Yosinski, Jeff Clune, Yoshua Bengio, and Hod Lipson. 2014. How transferable are features in deep neural networks?Advances in neural information processing systems 27 (2014).Google Scholar
Index Terms
- Streaming learning with Move-to-Data approach for image classification
Recommendations
Entropy-based Sampling for Streaming learning with Move-to-Data approach on Video
CBMI '23: Proceedings of the 20th International Conference on Content-based Multimedia IndexingThe current paradigm of training deep neural networks relies on large, annotated and representative datasets. They assume a static world where the target domain does not change. However, in the real-world, data changes over time and is often available ...
Incremental semi-supervised Extreme Learning Machine for Mixed data stream classification
AbstractWith an explosive growth of data generated in the Internet and other fields, the data stream classification has sparked broad interest recently. Nowadays, some of the challenges in data streams, such as concept drift detection and ...
Highlights- A soft one-hot encoding method for categorical data is designed.
- An incremental ...
Incremental Semi-Supervised classification of data streams via self-representative selection
An illustration of the proposed IS3RS approach.Display Omitted We advance an Incremental Semi-Supervised classification (ISSC) approach via Self-Representative Selection (IS3RS).We develop an incremental self-representative data selection strategy.Most ...
Comments