AI Paper Atlas

Research map / Architectures and efficiency / RNNs and normalization

ePC: Fast and Deep Predictive Coding in Digital Simulation

Cédric Goemaere, Gaspard Oliviers, Rafal Bogacz, Thomas Demeester

ICML 2026 · Architectures and efficiency › RNNs and normalization · approach: Hebbian and predictive coding · 2 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
reparameterize PC over prediction errors (ePC) instead of states (sPC)20-layer linear PC network · MNIST convergence to analytical optimumconvergence speed: over 100x fasterImproved
use ePC instead of sPC on deep non-linear MLPsPC network · Deep non-linear MLPssteps to convergence: sPC needed >100,000 stepsImproved
use ePC instead of sPC on VGG-7 with MSE lossVGG-7 · CIFAR-10accuracy (%): 77.79±0.34 -> 88.98±0.19Improved
use ePC instead of sPC on ResNet-18 with MSE lossResNet-18 · CIFAR-10accuracy (%): 53.74±0.43 -> 92.17±0.26Improved
use ePC instead of backprop on VGG-9 with MSE lossVGG-9 · CIFAR-100 (Top-1)accuracy (%): 66.28±0.29 -> 61.35±0.76Hurt
use ePC instead of backprop on small MLP4-layer MLP · FashionMNIST, MSE lossaccuracy (%): 88.79±0.21 -> 87.02±0.24Hurt
switch output loss from MSE to cross-entropy for ePCePC · VGG-7 CIFAR-100 (Top-1)accuracy (%): 66.55±0.45 -> 58.62±0.20Hurt
higher numerical precision (float64) for sPC (ablation)sPC · 20-layer MLP signal propagationsignal propagationNo clear effect
feedforward state initialization for sPCsPC · Signal propagationconvergenceMixed

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