AI Paper Atlas

Research map / Architectures and efficiency / Spiking networks

Biologically Inspired Dynamic Thresholds for Spiking Neural Networks

Jianchuan Ding, Bo Dong, Felix Heide, Yufei Ding, Yunduo Zhou, Baocai Yin, Xin Yang

NeurIPS 2022 · Architectures and efficiency › Spiking networks · approach: Bio-inspired spiking neuron models · 23 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
add BDETT dynamic thresholds to SAN-NRLIF-based SAN-NR · Obstacle avoidance, 0.2 degraded inputsuccess rate: 80% -> 90%Improved
add BDETT dynamic thresholds to SAN-NRSRM-based SAN-NR · Obstacle avoidance, GN weightsuccess rate: 37.2% ($\sigma\mbox{-}7.6$) -> 61.8% ($\sigma\mbox{-}2.9$)Improved
add BDETT dynamic thresholds to PopSANLIF-based PopSAN · HalfCheetah-v3, random joint positionreward: 7832 -> 8465Improved
add BDETT dynamic thresholds to PopSANLIF-based PopSAN · Ant-v3, GN weightreward: 637 -> 2782Improved
add BDETT dynamic thresholds to SCNNLIF-based SCNN · MNIST, FGSMTop-1 accuracy: 66.33% -> 69.14%Improved
add BDETT dynamic thresholds to SCNNLIF-based SCNN · MNIST, normalTop-1 accuracy: 99.42% -> 99.45%No clear effect
replace layerwise statistical cues with original fitted biological constants (ablation)BDETT (LIF) · Obstacle avoidance, standard conditionsuccess rate: 92.5% -> 0%Hurt
replace layerwise statistical cues with original fitted biological constants (ablation)BDETT (LIF) · HalfCheetah-v3, standard conditionreward: 11064 -> -35Hurt
replace layerwise statistical cues with original fitted biological constants (ablation)BDETT (LIF) · Ant-v3, standard conditionreward: 5276 -> -9Hurt
replace layerwise statistical cues with original fitted biological constants (ablation)BDETT (LIF) · MNIST, normalTop-1 accuracy: 99.45% -> 9.80%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