AI Paper Atlas

Research map / Architectures and efficiency / Spiking networks

Temporal-Coded Spiking Neural Networks with Dynamic Firing Threshold: Learning with Event-Driven Backpropagation

Wenjie Wei, Malu Zhang, Hong Qu, Ammar Belatreche, Jian Zhang, Hong Chen

ICCV 2023 · Architectures and efficiency › Spiking networks · approach: Training deep spiking networks · 38 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
use dynamic firing threshold with layer-dependent windows instead of fixed threshold (ablation)TTFS SNN (VGG11) · CIFAR-10accuracy: 10% -> 91.17%Improved
add DFT to reduce inactive neuronsTTFS SNN (VGG11) · CIFAR-10inactive neuron percentage per layerImproved
DTA-TTFS versus Zhang et al. 2020 on same MNIST networkDTA-TTFS · MNISTsparsity: 0.6614 -> 0.3913Improved
use DTA-TTFS VGG16 versus Zhou et al. 2021DTA-TTFS · CIFAR-10accuracy: 92.68% -> 93.05%Improved
DTA-TTFS versus Zhou et al. on spikesDTA-TTFS · CIFAR-10sparsity: 0.62 vs 0.2561Mixed
DTA-TTFS versus Park et al. 2021 on CIFAR-100DTA-TTFS · CIFAR-100accuracy: 65.98% -> 69.66%Improved
use DTA-TTFS short latency coding versus TTFS [38]DTA-TTFS · CIFAR-10 energy (TrueNorth)normalized energy: 0.6478 -> 0.1987Improved
deploy DTA-TTFS on CanMore versus Esser et al.DTA-TTFS · CIFAR-10 hardware simulationenergy (uJ/frame): 163.65 -> 50.36Improved
DTA-TTFS VGG11 versus rate-coded conversionDTA-TTFS · CIFAR-10accuracy: 91.24% vs 91.17%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