AI Paper Atlas

Research map / Architectures and efficiency / Spiking networks

Developmental Plasticity-Inspired Adaptive Pruning for Deep Spiking and Artificial Neural Networks

Bing Han, Feifei Zhao, Yi Zeng, Guobin Shen

TPAMI 2024 · Architectures and efficiency › Spiking networks · approach: Training deep spiking networks · 21 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 BCM synaptic plasticity with Hebb or STDP (ablation)DPAP · SNN pruning, several datasetstest accuracyImproved
add DPAP pruning to DNNDNN · MNIST (DNN)epochs to reach 99.20% accuracy: 50 -> 39 epochsImproved
add DPAP pruning to DNNDNN · CIFAR-10 (DNN)convergence speed: 2.84xImproved
add temporal pruning to spatial pruningDPAP · DVS-Gesture (SNN)accuracy: 97.83% -> 98.20%Improved
add DPAP pruning to SNNSNN · MNIST, N-MNIST, DVS-Gestureaccuracy: ~0.31% average gain at ~63% compressionImproved
increase pruning rate beyond about 60% (ablation)DPAP SNN · spatial and temporal datasetsaccuracy: peak at about 60%Mixed
increase DNN pruning rate on CIFAR-10 (ablation)DPAP DNN · CIFAR-10 (DNN)accuracy: 94.76% at 41.54%Mixed
DPAP versus Grad R at 75% sparsityDPAP · DVS-Gesture (SNN)accuracy: 91.95% -> 97.83%Improved
DPAP versus ThiNet at similar pruning rateDPAP · ImageNet (DNN)accuracy: +2.25%Improved
DPAP versus HRank at similar pruning rateDPAP · ImageNet (DNN)accuracy: +1.28%Improved
prune ImageNet SNN with DPAP at 51.71% sparsity (ablation)DPAP · ImageNet (SNN)accuracy: 63.74% -> 60.41%Hurt

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