AI Paper Atlas

Research map / Architectures and efficiency / Spiking networks

Efficient Training of Spiking Neural Networks with Multi-Parallel Implicit Stream Architecture

Zhigao Cao, Meng Li, Xiashuang Wang, Haoyu Wang, Fan Wang, Youjun Li, Zigang Huang

ECCV 2024 · Architectures and efficiency › Spiking networks · approach: Training deep spiking networks · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
replace BPTT with implicit equilibrium training (MPIS) on Fashion-MNIST at T=30SNN 133K · Fashion-MNISTaccuracy: 89.60% -> 93.14%Improved
replace BPTT with MPIS at T=100 on Fashion-MNISTSNN 133K · Fashion-MNISTtraining memory: 4.8G -> 1.2GImproved
replace BPTT with MPIS on N-MNIST at T=30SNN 213K · N-MNISTtraining memory: 12.8G -> 3.3GImproved
replace BPTT with MPISSNN on N-MNIST · N-MNISTfiring rate (layer 1): 5.06e-2 -> 8.0e-4Improved
replace IDE-Nets with multi-parallel implicit streams at 30 time stepsCIFAR-10 11.8M · CIFAR-10accuracy: 90.37% -> 92.79%Improved
replace IDE-Nets with MPIS at 30 time stepsCIFAR-10 11.8M · CIFAR-10training time: 12min10s -> 2min34sImproved
replace IDE-Nets with MPIS at 30 time stepsCIFAR-100 14.8M · CIFAR-100accuracy: 70.26% -> 73.19%Improved
scale MPIS parameters to about twice IDE-Net size (ablation)MPIS-SNNs · CIFAR-100accuracy: 73.19% -> 74.40%Improved
use two parallel implicit streams instead of a single stream (ablation)MPIS-SNNs · N-MNIST, CIFAR-10, CIFAR-100convergence speedImproved
increase time steps from 30 to 100 for IDE-Nets (ablation)IDE-Nets · CIFAR-10accuracy: 90.37% -> 90.57%No clear effect
increase time steps from 30 to 100 for MPIS on Fashion-MNIST (ablation)MPIS · Fashion-MNISTaccuracy: 93.14% -> 93.23%No clear effect

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy