AI Paper Atlas

Research map / Architectures and efficiency / RNNs and normalization

K-prop: Deep Online Learning by Backpropagating Temporal Kernels

Bingkun Liu, Paul Haider, Federico Benitez, Mihai A. Petrovici, Walter Senn

NeurIPS 2026 · Architectures and efficiency › RNNs and normalization · approach: Brain-inspired recurrent architectures · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Proposes K-prop, an online learning method for leaky (self-recurrent) neural networks that backpropagates through temporal kernels to handle complex temporal gradient dependencies.

Problem

Efficient and effective online training of self-recurrent (leaky integration) neural networks is lacking because the temporal dependencies of the gradient are complex.

Approach

Backpropagates gradients through temporal kernels of leaky neurons to enable online learning in leaky neural networks, aiming to avoid the complex temporal gradient dependencies that hinder existing online methods.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Use K-prop instead of OSTL for online trainingleaky network · Google Speech Commands (20 commands)test accuracy (%): 90.61 -> 91.52Improved
Use K-prop instead of OSTL for online trainingleaky network Lk4 · MNIST1D (Lk4)test accuracy (%): 96.08 -> 96.53Improved
Use K-prop instead of OSTL for online trainingleaky network Lk5 · MNIST1D (Lk5)test accuracy (%): 97.03 -> 97.56Improved
Replace exact BPTT with K-propleaky network · Google Speech Commands (20 commands)test accuracy (%): 92.47 -> 91.52Hurt
Freeze progressively more early layers during training (ablation)K-prop Lk5 · MNIST1D (Lk5)accuracyHurt
Use low-pass-filtered rho' as surrogate instead of instantaneous rho' (ablation)K-prop · gradient cosine dissimilarity, random networks on MNIST1Dcosine dissimilarity to BPTTImproved
Use sigmoid instead of ReLU or tanh activations (ablation)K-prop · gradient cosine dissimilarity, random networks on MNIST1Dcosine dissimilarity to BPTTImproved
Increase network depth (distance from output layer) (ablation)K-prop gradient · gradient cosine dissimilarity, random networks on MNIST1Dcosine dissimilarity to BPTTHurt
Increase neuron time constant tau (ablation)K-prop gradient · gradient cosine dissimilarity, random networks on MNIST1Dcosine dissimilarity to BPTTMixed
Apply weight updates at discrete intervals instead of every step (ablation)online learning · N-sum / sequence taskslearning stabilityImproved
Add Gaussian noise to eligibility traces Z (ablation)K-prop Lk4 · MNIST1Daccuracy (%): 96.53 -> 94.5 (sigma=1), 92.3 (sigma=10)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