Summary of Neuromoco: a Neuromorphic Momentum Contrast Learning Method For Spiking Neural Networks, by Yuqi Ma and Huamin Wang and Hangchi Shen and Xuemei Chen and Shukai Duan and Shiping Wen
NeuroMoCo: A Neuromorphic Momentum Contrast Learning Method for Spiking Neural Networks
by Yuqi Ma, Huamin Wang, Hangchi Shen, Xuemei Chen, Shukai Duan, Shiping Wen
First submitted to arxiv on: 10 Jun 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Artificial Intelligence (cs.AI)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary A novel approach, Neuromorphic Momentum Contrast Learning (NeuroMoCo), is introduced to improve the performance of brain-inspired spiking neural networks (SNNs) on event-based neuromorphic datasets. This method extends self-supervised pre-training benefits to SNNs and effectively stimulates their potential. NeuroMoCo combines momentum contrastive learning with a novel loss function, MixInfoNCE, tailored to temporal characteristics. The approach is evaluated through ablation experiments and achieves state-of-the-art (SOTA) benchmarks on DVS-CIFAR10, DVS128Gesture, and N-Caltech101 datasets. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Brain-inspired spiking neural networks are being studied because they can understand events happening at different times. This makes them good for handling special kinds of data that have time series information. The problem is that these datasets are very hard to analyze because of their unique characteristics. To solve this, a new way of training SNNs called NeuroMoCo was developed. It uses self-supervised learning to make the networks better at understanding events. This approach achieved state-of-the-art results on several datasets. |
Keywords
» Artificial intelligence » Loss function » Self supervised » Time series
