Self-cross Feature based Spiking Neural Networks for Efficient Few-shot Learning

Qi Xu, Junyang Zhu, Dongdong Zhou, Hao Chen, Yang Liu, Jiangrong Shen, Qiang Zhang
Proceedings of the 42nd International Conference on Machine Learning, PMLR 267:70007-70019, 2025.

Abstract

Deep neural networks (DNNs) excel in computer vision tasks, especially, few-shot learning (FSL), which is increasingly important for generalizing from limited examples. However, DNNs are computationally expensive with scalability issues in real world. Spiking Neural Networks (SNNs), with their event-driven nature and low energy consumption, are particularly efficient in processing sparse and dynamic data, though they still encounter difficulties in capturing complex spatiotemporal features and performing accurate cross-class comparisons. To further enhance the performance and efficiency of SNNs in few-shot learning, we propose a few-shot learning framework based on SNNs, which combines a self-feature extractor module and a cross-feature contrastive module to refine feature representation and reduce power consumption. We apply the combination of temporal efficient training loss and InfoNCE loss to optimize the temporal dynamics of spike trains and enhance the discriminative power. Experimental results show that the proposed FSL-SNN significantly improves the classification performance on the neuromorphic dataset N-Omniglot, and also achieves competitive performance to ANNs on static datasets such as CUB and miniImageNet with low power consumption.

Cite this Paper

BibTeX
@InProceedings{pmlr-v267-xu25aq, title = {Self-cross Feature based Spiking Neural Networks for Efficient Few-shot Learning}, author = {Xu, Qi and Zhu, Junyang and Zhou, Dongdong and Chen, Hao and Liu, Yang and Shen, Jiangrong and Zhang, Qiang}, booktitle = {Proceedings of the 42nd International Conference on Machine Learning}, pages = {70007--70019}, year = {2025}, editor = {Singh, Aarti and Fazel, Maryam and Hsu, Daniel and Lacoste-Julien, Simon and Berkenkamp, Felix and Maharaj, Tegan and Wagstaff, Kiri and Zhu, Jerry}, volume = {267}, series = {Proceedings of Machine Learning Research}, month = {13--19 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v267/main/assets/xu25aq/xu25aq.pdf}, url = {https://proceedings.mlr.press/v267/xu25aq.html}, abstract = {Deep neural networks (DNNs) excel in computer vision tasks, especially, few-shot learning (FSL), which is increasingly important for generalizing from limited examples. However, DNNs are computationally expensive with scalability issues in real world. Spiking Neural Networks (SNNs), with their event-driven nature and low energy consumption, are particularly efficient in processing sparse and dynamic data, though they still encounter difficulties in capturing complex spatiotemporal features and performing accurate cross-class comparisons. To further enhance the performance and efficiency of SNNs in few-shot learning, we propose a few-shot learning framework based on SNNs, which combines a self-feature extractor module and a cross-feature contrastive module to refine feature representation and reduce power consumption. We apply the combination of temporal efficient training loss and InfoNCE loss to optimize the temporal dynamics of spike trains and enhance the discriminative power. Experimental results show that the proposed FSL-SNN significantly improves the classification performance on the neuromorphic dataset N-Omniglot, and also achieves competitive performance to ANNs on static datasets such as CUB and miniImageNet with low power consumption.} }
Endnote
%0 Conference Paper %T Self-cross Feature based Spiking Neural Networks for Efficient Few-shot Learning %A Qi Xu %A Junyang Zhu %A Dongdong Zhou %A Hao Chen %A Yang Liu %A Jiangrong Shen %A Qiang Zhang %B Proceedings of the 42nd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2025 %E Aarti Singh %E Maryam Fazel %E Daniel Hsu %E Simon Lacoste-Julien %E Felix Berkenkamp %E Tegan Maharaj %E Kiri Wagstaff %E Jerry Zhu %F pmlr-v267-xu25aq %I PMLR %P 70007--70019 %U https://proceedings.mlr.press/v267/xu25aq.html %V 267 %X Deep neural networks (DNNs) excel in computer vision tasks, especially, few-shot learning (FSL), which is increasingly important for generalizing from limited examples. However, DNNs are computationally expensive with scalability issues in real world. Spiking Neural Networks (SNNs), with their event-driven nature and low energy consumption, are particularly efficient in processing sparse and dynamic data, though they still encounter difficulties in capturing complex spatiotemporal features and performing accurate cross-class comparisons. To further enhance the performance and efficiency of SNNs in few-shot learning, we propose a few-shot learning framework based on SNNs, which combines a self-feature extractor module and a cross-feature contrastive module to refine feature representation and reduce power consumption. We apply the combination of temporal efficient training loss and InfoNCE loss to optimize the temporal dynamics of spike trains and enhance the discriminative power. Experimental results show that the proposed FSL-SNN significantly improves the classification performance on the neuromorphic dataset N-Omniglot, and also achieves competitive performance to ANNs on static datasets such as CUB and miniImageNet with low power consumption.
APA
Xu, Q., Zhu, J., Zhou, D., Chen, H., Liu, Y., Shen, J. & Zhang, Q.. (2025). Self-cross Feature based Spiking Neural Networks for Efficient Few-shot Learning. Proceedings of the 42nd International Conference on Machine Learning, in Proceedings of Machine Learning Research 267:70007-70019 Available from https://proceedings.mlr.press/v267/xu25aq.html.

Related Material