Publications

You can also find my articles on my Google Scholar profile.

High-Performance Temporal Reversible Spiking Neural Networks with O(L) Training Memory and O(1) Inference Cost

Published in ICML 2024 Spotlight, 2024

Multi-timestep simulation of brain-inspired Spiking Neural Networks (SNNs) boost memory requirements during training and increase inference energy cost. Current training methods cannot simultaneously solve both training and inference dilemmas. This work proposes a novel Temporal Reversible architecture for SNNs (T-RevSNN) to jointly address the training and inference challenges by altering the forward propagation of SNNs. We turn off the temporal dynamics of most spiking neurons and design multi-level temporal reversible interactions at temporal turn-on spiking neurons, resulting in a O(L) training memory. Combined with the temporal reversible nature, we redesign the input encoding and network organization of SNNs to achieve O(1) inference energy cost. Then, we finely adjust the internal units and residual connections of the basic SNN block to ensure the effectiveness of sparse temporal information interaction. T-RevSNN achieves excellent accuracy on ImageNet, while the memory efficiency, training time acceleration, and inference energy efficiency can be significantly improved by 8.6×, 2.0×, and 1.6×, respectively. This work is expected to break the technical bottleneck of significantly increasing memory cost and training time for large-scale SNNs while maintaining high performance and low inference energy cost.

Recommended citation: Hu J K, Yao M, Qiu X, et al. High-Performance Temporal Reversible Spiking Neural Networks with O(L) Training Memory and O(1) Inference Cost. ICML 2024.
Download Paper

RSC-SNN: Exploring the Trade-off Between Adversarial Robustness and Accuracy in Spiking Neural Networks via Randomized Smoothing Coding

Published in ACM MM 2024, 2024

Spiking Neural Networks (SNNs) have received widespread attention due to their unique neuronal dynamics and low-power nature. Previous research empirically shows that SNNs with Poisson coding are more robust than Artificial Neural Networks (ANNs) on small-scale datasets. However, it is still unclear in theory how the adversarial robustness of SNNs is derived, and whether SNNs can still maintain its adversarial robustness advantage on large-scale dataset tasks. This work theoretically demonstrates that SNN’s inherent adversarial robustness stems from its Poisson coding. We reveal the conceptual equivalence of Poisson coding and randomized smoothing in defense strategies, and analyze in depth the trade-off between accuracy and adversarial robustness in SNNs via the proposed Randomized Smoothing Coding (RSC) method. Experiments demonstrate that the proposed RSC-SNNs show remarkable adversarial robustness, surpassing ANNs and achieving state-of-the-art robustness results on large-scale dataset ImageNet.

Recommended citation: Wu K, Yao M, Chou Y, Qiu X, et al. RSC-SNN: Exploring the Trade-off Between Adversarial Robustness and Accuracy in Spiking Neural Networks via Randomized Smoothing Coding. ACM MM 2024.
Download Paper

Tensor decomposition based attention module for spiking neural networks

Published in Knowledge-Based Systems, 2024

The attention mechanism has been proven to be an effective way to improve the performance of spiking neural networks (SNNs). However, from the perspective of tensor decomposition to examine the existing attention modules, we find that the rank of the attention maps generated by previous methods is fixed at 1, lacking the flexibility to adjust for specific tasks. To tackle this problem, we propose an attention module, namely Projected-full Attention (PFA), where the rank of the generated attention maps can be determined based on the characteristics of different tasks. Additionally, the parameter count of PFA grows linearly with the data scale. PFA is composed of the linear projection of spike tensor (LPST) module and attention map composing (AMC) module. In LPST, we start by compressing the original spike tensor into three projected tensors with learnable parameters for each dimension. Then, in AMC, we exploit the inverse procedure of the tensor decomposition process to combine the three tensors into the attention map using a so-called connecting factor. To validate the effectiveness of the proposed PFA module, we integrate it into the widely used VGG and ResNet architectures for classification tasks. Our method achieves state-of-the-art performance on both static and dynamic benchmark datasets, surpassing the existing SNN models with Transformer-based and CNN-based backbones.

Recommended citation: Deng H, Zhu R, Qiu X, et al. Tensor decomposition based attention module for spiking neural networks[J]. Knowledge-Based Systems, 2024, 295: 111780.
Download Paper

Gated attention coding for training high-performance and efficient spiking neural networks

Published in AAAI 2024 Poster, 2024

Spiking neural networks (SNNs) are emerging as an energy-efficient alternative to traditional artificial neural networks (ANNs) due to their unique spike-based event-driven nature. Coding is crucial in SNNs as it converts external input stimuli into spatio-temporal feature sequences. However, most existing deep SNNs rely on direct coding that generates powerless spike representation and lacks the temporal dynamics inherent in human vision. Hence, we introduce Gated Attention Coding (GAC), a plug-and-play module that leverages the multi-dimensional gated attention unit to efficiently encode inputs into powerful representations before feeding them into the SNN architecture. GAC functions as a preprocessing layer that does not disrupt the spike-driven nature of the SNN, making it amenable to efficient neuromorphic hardware implementation with minimal modifications. Through an observer model theoretical analysis, we demonstrate GAC’s attention mechanism improves temporal dynamics and coding efficiency. Experiments on CIFAR10/100 and ImageNet datasets demonstrate that GAC achieves state-of-the-art accuracy with remarkable efficiency. Notably, we improve top-1 accuracy by 3.10% on CIFAR100 with only 6-time steps and 1.07% on ImageNet while reducing energy usage to 66.9% of the previous works. To our best knowledge, it is the first time to explore the attention-based dynamic coding scheme in deep SNNs, with exceptional effectiveness and efficiency on large-scale datasets.

Recommended citation: Qiu X, Zhu R J, Chou Y, et al. Gated attention coding for training high-performance and efficient spiking neural networks. AAAI 2024.
Download Paper

When Spiking Neural Networks Meet Temporal Attention Image Decoding and Adaptive Spiking Neuron

Published in ICLR 2024 TinyPaper, 2023

Spiking Neural Networks (SNNs) are capable of encoding and processing temporal information in a biologically plausible way. However, most existing SNN-based methods for image tasks do not fully exploit this feature. Moreover, they often overlook the role of adaptive threshold in spiking neurons, which can enhance their dynamic behavior and learning ability. To address these issues, we propose a novel method for image decoding based on temporal attention (TAID) and an adaptive Leaky-Integrate-and-Fire (ALIF) neuron model. Our method leverages the temporal information of SNN outputs to generate high-quality images that surpass the state-of-the-art (SOTA) in terms of Inception score, Fréchet Inception Distance, and Fréchet Autoencoder Distance. Furthermore, our ALIF neuron model achieves remarkable classification accuracy on MNIST (99.78\%) and CIFAR-10 (93.89\%) datasets, demonstrating the effectiveness of learning adaptive thresholds for spiking neurons.

Recommended citation: Qiu X, Luan Z, Wang Z, et al. When Spiking neural networks meet temporal attention image decoding and adaptive spiking neuron. ICLR 2023.
Download Paper

VTSNN: a virtual temporal spiking neural network

Published in Frontiers in Neuroscience, 2023

Spiking neural networks (SNNs) have recently demonstrated outstanding performance in a variety of high-level tasks, such as image classification. However, advancements in the field of low-level assignments, such as image reconstruction, are rare. This may be due to the lack of promising image encoding techniques and corresponding neuromorphic devices designed specifically for SNN-based low-level vision problems. This paper begins by proposing a simple yet effective undistorted weighted-encoding-decoding technique, which primarily consists of an Undistorted Weighted-Encoding (UWE) and an Undistorted Weighted-Decoding (UWD). The former aims to convert a gray image into spike sequences for effective SNN learning, while the latter converts spike sequences back into images. Then, we design a new SNN training strategy, known as Independent-Temporal Backpropagation (ITBP) to avoid complex loss propagation in spatial and temporal dimensions, and experiments show that ITBP is superior to Spatio-Temporal Backpropagation (STBP). Finally, a so-called Virtual Temporal SNN (VTSNN) is formulated by incorporating the above-mentioned approaches into U-net network architecture, fully utilizing the potent multiscale representation capability. Experimental results on several commonly used datasets such as MNIST, F-MNIST, and CIFAR10 demonstrate that the proposed method produces competitive noise-removal performance extremely which is superior to the existing work. Compared to ANN with the same architecture, VTSNN has a greater chance of achieving superiority while consuming ~1/274 of the energy. Specifically, using the given encoding-decoding strategy, a simple neuromorphic circuit could be easily constructed to maximize this low-carbon strategy.

Recommended citation: Qiu X R, Wang Z R, Luan Z, et al. VTSNN: a virtual temporal spiking neural network. Frontiers in Neuroscience, 2023, 17: 1091097.
Download Paper