AI Paper Atlas

Research map / Probabilistic and scientific modeling / Dynamical systems

UltraLIF: Fully Differentiable Spiking Neural Networks via Ultradiscretization and Max-Plus Algebra

Jose Marie Antonio Miñoza

ICML 2026 · Probabilistic and scientific modeling › Dynamical systems · approach: Physics-informed dynamics learning · 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 surrogate-gradient spikes with ultradiscretized (LSE) neuronsUltraLIF family · SHD (T=1)accuracy (%): 40.02 -> 51.24Improved
replace surrogate-gradient spikes with ultradiscretized neuronsUltraLIF family · MNIST (T=1)accuracy (%): 95.58 -> 95.67No clear effect
increase timesteps to T>=10UltraLIF family vs surrogate baselines · SHD, N-MNISTaccuracy (%)Hurt
add sparsity penalty lambda=0.1 (ablation)UltraPLIF · CIFAR-10 (T=1)accuracy (%): 43.27 -> 43.60 (spike rate 0.458 -> 0.240)Improved
add sparsity penalty lambda=0.1 (ablation)UltraPLIF · CIFAR-10 (T=30)accuracy (%): 46.58 -> 46.98Improved
add sparsity penalty lambda=0.1 (ablation)UltraDPLIF · CIFAR-10 (T=10)accuracy (%): 45.75 -> 45.32Hurt
use hard binary spikes at inference (ablation)UltraLIF family · MNIST (T=1)accuracy (%): 95.67 -> 93.72Hurt
use hard binary spikes at inference (ablation)UltraLIF family · CIFAR-10 (T=1)accuracy (%): 43.27 -> 43.35No clear effect
learnable temperature epsilon replaces fixed surrogate sharpnessUltraPLIF on ResNet50 spiking head · CIFAR-10 (ResNet50, T=1)accuracy (%): 31.83 -> 92.78Improved
use ultradiscretized neurons in a 2-layer conv SNNUltraDPLIF · CIFAR-10 (Conv 2L, T=1)accuracy (%): 74.37 -> 70.54Hurt
use ultradiscretized neurons in ResNet18 spiking headUltraDPLIF · CIFAR-10 (ResNet18, T=1)accuracy (%): 93.12 -> 93.37No clear effect
use ultradiscretized neurons in fully connected nets of depth 1 to 3UltraPLIF/UltraDPLIF · CIFAR-10 (FC 1L)accuracy (%): 41.79 -> 46.40Improved

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