A Graph Meta-Network for Learning on Kolmogorov-Arnold Networks

• URL: http://arxiv.org/abs/2602.16316v1
• Date: Wed, 18 Feb 2026 09:53:53 GMT
• Title: A Graph Meta-Network for Learning on Kolmogorov-Arnold Networks
• Authors: Guy Bar-Shalom, Ami Tavory, Itay Evron, Maya Bechler-Speicher, Ido Guy, Haggai Maron,
• Abstract summary: We develop WS-KAN, the first weight-space architecture that learns on Kolmogorov-Arnold Networks (KANs)<n>We show that KANs share the same permutation symmetries asArnolds, and propose the KAN-graph.<n>We analyze WS-KAN's expressive power, showing it can replicate an input KAN's forward pass.
• Score: 28.337888387996156
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Weight-space models learn directly from the parameters of neural networks, enabling tasks such as predicting their accuracy on new datasets. Naive methods -- like applying MLPs to flattened parameters -- perform poorly, making the design of better weight-space architectures a central challenge. While prior work leveraged permutation symmetries in standard networks to guide such designs, no analogous analysis or tailored architecture yet exists for Kolmogorov-Arnold Networks (KANs). In this work, we show that KANs share the same permutation symmetries as MLPs, and propose the KAN-graph, a graph representation of their computation. Building on this, we develop WS-KAN, the first weight-space architecture that learns on KANs, which naturally accounts for their symmetry. We analyze WS-KAN's expressive power, showing it can replicate an input KAN's forward pass - a standard approach for assessing expressiveness in weight-space architectures. We construct a comprehensive ``zoo'' of trained KANs spanning diverse tasks, which we use as benchmarks to empirically evaluate WS-KAN. Across all tasks, WS-KAN consistently outperforms structure-agnostic baselines, often by a substantial margin. Our code is available at https://github.com/BarSGuy/KAN-Graph-Metanetwork.

Related papers

• NNiT: Width-Agnostic Neural Network Generation with Structurally Aligned Weight Spaces [15.631276865948097]
  We introduce Neural Network Diffusion Transformers (NNiTs), which generate weights in a width-agnostic manner.<n>On ManiSkill3 robotics tasks, NNiT achieves >85% success on architecture topologies unseen during training, while baseline approaches fail to generalize.
  arXiv  Detail & Related papers  (2026-02-26T20:47:30Z)
• Automatic Grid Updates for Kolmogorov-Arnold Networks using Layer Histograms [0.6445605125467574]
  Kolmogorov-Arnold Networks (KANs) are a class of neural networks that have received increased attention in recent literature.<n>KANs leverage parameterized, trainable activation functions and offer several benefits including improved interpretability and higher accuracy on learning symbolic equations.
  arXiv  Detail & Related papers  (2025-11-11T18:51:44Z)
• RSQ: Learning from Important Tokens Leads to Better Quantized LLMs [65.5558181902098]
  Layer-wise quantization is a key technique for efficiently compressing large models without expensive retraining.<n>We propose RSQ (Rotate, Scale, then Quantize), which applies rotations to the model to mitigate outliers.<n>We demonstrate that RSQ consistently outperforms baseline methods across multiple downstream tasks and three model families.
  arXiv  Detail & Related papers  (2025-03-03T18:46:33Z)
• Incorporating Arbitrary Matrix Group Equivariance into KANs [69.30866522377694]
  Kolmogorov-Arnold Networks (KANs) have seen great success in scientific domains.<n>We propose Equivariant Kolmogorov-Arnold Networks (EKAN) to broaden their applicability to more fields.
  arXiv  Detail & Related papers  (2024-10-01T06:34:58Z)
• GKAN: Graph Kolmogorov-Arnold Networks [6.267574471145217]
  We introduce Graph Kolmogorov-Arnold Networks (GKAN) GKAN is an innovative neural network architecture that extends the principles of the recently proposed Kolmogorov Networks (KAN) to graphstructured data. We evaluate GKAN empirically using a semi-supervised graph learning task on a real-world dataset.
  arXiv  Detail & Related papers  (2024-06-10T17:09:38Z)
• Set-based Neural Network Encoding Without Weight Tying [91.37161634310819]
  We propose a neural network weight encoding method for network property prediction.<n>Our approach is capable of encoding neural networks in a model zoo of mixed architecture.<n>We introduce two new tasks for neural network property prediction: cross-dataset and cross-architecture.
  arXiv  Detail & Related papers  (2023-05-26T04:34:28Z)
• Equivariant Architectures for Learning in Deep Weight Spaces [54.61765488960555]
  We present a novel network architecture for learning in deep weight spaces. It takes as input a concatenation of weights and biases of a pre-trainedvariant. We show how these layers can be implemented using three basic operations.
  arXiv  Detail & Related papers  (2023-01-30T10:50:33Z)
• Pre-Trained Models for Heterogeneous Information Networks [57.78194356302626]
  We propose a self-supervised pre-training and fine-tuning framework, PF-HIN, to capture the features of a heterogeneous information network. PF-HIN consistently and significantly outperforms state-of-the-art alternatives on each of these tasks, on four datasets.
  arXiv  Detail & Related papers  (2020-07-07T03:36:28Z)
• Fitting the Search Space of Weight-sharing NAS with Graph Convolutional Networks [100.14670789581811]
  We train a graph convolutional network to fit the performance of sampled sub-networks. With this strategy, we achieve a higher rank correlation coefficient in the selected set of candidates.
  arXiv  Detail & Related papers  (2020-04-17T19:12:39Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.