Advancing Spiking Neural Networks Towards Multiscale Spatiotemporal Interaction Learning
Authors: Yimeng Shan, Malu Zhang, Rui-jie Zhu, Xuerui Qiu, Jason K. Eshraghian, Haicheng Qu
AAAI 2025 | Venue PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Our approach has achieved state-of-the-art results on mainstream neuromorphic datasets. Additionally, we have reached a performance of 77.1% on the Imagenet-1K dataset using a 104-layer Res Net architecture enhanced with SMA and AZO. This achievement confirms the state-of-the-art performance of SNNs with non-transformer architectures and underscores the effectiveness of our method in bridging the performance gap between SNN models and traditional ANN models. |
| Researcher Affiliation | Academia | 1Liaoning Technical University, China 2University of Electronic Science and Technology of China, China 3University of California, Santa Cruz, USA |
| Pseudocode | Yes | Algorithm 1: Attention Zoneout |
| Open Source Code | Yes | Code https://github.com/Ym Shan/SMA-AZO |
| Open Datasets | Yes | Even with the use of spiking coding, the spatiotemporal interaction in static image datasets remains limited. Therefore, we only evaluated the classification performance of our proposed SMA-SNN and SMA-AZO-SNN architectures on three prominent neuromorphic datasets (DVS128 Gesture (Amir et al. 2017), CIFAR10-DVS (Li et al. 2017), and N-Caltech101 (Orchard et al. 2015)) as well as the Image Net-1K dataset (Deng et al. 2009). |
| Dataset Splits | No | The paper does not provide specific dataset split information needed to reproduce the data partitioning. It mentions using several known datasets but does not detail how these datasets were split into training, validation, or test sets, nor does it refer to specific standard splits with citations for reproducibility. |
| Hardware Specification | No | The paper does not provide specific hardware details (exact GPU/CPU models, processor types with speeds, memory amounts, or detailed computer specifications) used for running its experiments. |
| Software Dependencies | No | The paper does not provide specific ancillary software details (e.g., library or solver names with version numbers like Python 3.8, CPLEX 12.4) needed to replicate the experiment. |
| Experiment Setup | No | All network structures and hyperparameter settings utilized in the experiment are detailed in Sec. A of Supplementary Material. |