AI Paper Atlas

Research map / Architectures and efficiency / RNNs and normalization

Initialization Schemes for Kolmogorov–Arnold Networks: An Empirical Study

Spyros Rigas, Dhruv Verma, Georgios Alexandridis, Yixuan Wang

ICLR 2026 · Architectures and efficiency › RNNs and normalization · approach: Theory of depth and width · 6 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
use power-law initialization instead of baselinespline KAN · function fittingruns improving loss and L2 vs baseline: 94.79%-100.00% of runsImproved
use power-law initialization instead of baselinespline KAN · forward PDEsruns improving loss and L2 vs baseline: 73.61%-94.44% of runsImproved
use Glorot-inspired initialization instead of baselinespline KAN · function fittingruns improving loss and L2 vs baseline: 60.41%-78.13% of runsImproved
use Glorot-inspired initialization instead of baselinespline KAN · forward PDEsruns improving loss and L2 vs baseline: 36.11%-62.50% of runsMixed
use LeCun-numerical initialization instead of baselinespline KAN · function fittingruns improving loss and L2 vs baseline: 0.00%-8.33% of runsHurt
use LeCun-normalized initialization instead of baselinespline KAN · forward PDEsruns improving loss and L2 vs baseline: 0.00% of runsHurt
use normalized spline basis (LeCun-normalized) instead of numerical estimatespline KAN · function fittingruns improving loss and L2 vs baseline: e.g. 0.00 -> 5.21 (f2)Improved
use normalized spline basis (LeCun-normalized) instead of numerical estimatespline KAN · forward PDEsruns improving loss and L2 vs baseline: 6.94 -> 0.00 (Burgers)Hurt
fix power-law exponents at alpha=0.25, beta=1.0 rather than tuning per setting (ablation)power-law init · forward PDEsruns improving loss and L2 vs baseline: 94.44 -> 83.33 (Allen-Cahn); 73.61 -> 54.93 (Burgers)Mixed
use power-law initialization (alpha=0.25, beta=1.75) instead of baselinespline KAN, large architecture · Feynman subsetfinal training loss and L2Improved
use baseline instead of Glorot/power-law when scaling from small to large architecturespline KAN · Feynman subsetfinal lossHurt

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