mlsanityData to 7 Oct 2026

SigKAN: Signature-Weighted Kolmogorov-Arnold Networks for Time Series

arXiv 2406.17890 ↗First posted 25 Jun 2024cs.LGcited by 13

Not on the map. Why?

Abstract (arXiv)

We propose a novel approach that enhances multivariate function approximation using learnable path signatures and Kolmogorov-Arnold networks (KANs). We enhance the learning capabilities of these networks by weighting the values obtained by KANs using learnable path signatures, which capture important geometric features of paths. This combination allows for a more comprehensive and flexible representation of sequential and temporal data. We demonstrate through studies that our SigKANs with learnable path signatures perform better than conventional methods across a range of function approximation challenges. By leveraging path signatures in neural networks, this method offers intriguing opportunities to enhance performance in time series analysis and time series forecasting, among other fields.

Key facts. Cited by 13 papers in the map (data through 7 Oct 2026). Builds on 16 of its 36 references that are in the map. First posted on arXiv 25 Jun 2024.

Builds on

16 of 36 references.

References · 16

  1. 2017Attention Is All You Needcited by 59,932
  2. 2016Layer Normalizationcited by 5,579
  3. 2015Fast and Accurate Deep Network Learning by Exponential Linear Units (ELUs)cited by 1,061
  4. 2017Sigmoid-Weighted Linear Units for Neural Network Function Approximation in Reinforcement Learningcited by 935
  5. 2019Temporal Fusion Transformers for Interpretable Multi-horizon Time Series Forecastingcited by 748
  6. 2016Language Modeling with Gated Convolutional Networkscited by 688
  7. 2024KAN: Kolmogorov-Arnold Networkscited by 634
  8. 2024Wav-KAN: Wavelet Kolmogorov-Arnold Networkscited by 118
  9. 2016A Primer on the Signature Method in Machine Learningcited by 100
  10. 2024Kolmogorov-Arnold Networks (KANs) for Time Series Analysiscited by 97
  11. 2024TKAN: Temporal Kolmogorov-Arnold Networkscited by 70
  12. 2024A Temporal Kolmogorov-Arnold Transformer for Time Series Forecastingcited by 50
4 more
  1. 2017Autoregressive Convolutional Neural Networks for Asynchronous Time Seriescited by 31
  2. 2019Embedding and learning with signaturescited by 29
  3. 2017Developing the Path Signature Methodology and its Application to Landmark-based Human Action Recognitioncited by 20
  4. 2017A signature-based machine learning model for bipolar disorder and borderline personality disordercited by 15

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 6; 2025: 5; 2026: 2.

13 papers.

Sort:

By area:

  1. 2024A Survey on Kolmogorov-Arnold Networkcited by 67
  2. 2024PowerMLP: An Efficient Version of KANcited by 14
  3. 2024Exploring Kolmogorov-Arnold Networks for Interpretable Time Series Classificationcited by 8
  4. 2024A Gated Residual Kolmogorov-Arnold Networks for Mixtures of Expertscited by 6
  5. 2024A Temporal Linear Network for Time Series Forecastingcited by 4
  6. 2025Deep Learning for VWAP Execution in Crypto Markets: Beyond the Volume Curvecited by 4
  7. 2025SigGate: Enhancing Recurrent Neural Networks with Signature-Based Gating Mechanismscited by 3
  8. 2024Kolmogorov Arnold Networks in Fraud Detection: Bridging the Gap Between Theory and Practicecited by 2
  9. 2025Recurrent Neural Networks for Dynamic VWAP Execution: Adaptive Trading Strategies with Temporal Kolmogorov-Arnold Networkscited by 2
  10. 2025VWAP Execution with Signature-Enhanced Transformers: A Multi-Asset Learning Approachcited by 1
  11. 2026Hardware-Oriented Inference Complexity of Kolmogorov-Arnold Networkscited by 1
  12. 2025HKAN: Hierarchical Kolmogorov-Arnold Network without Backpropagationcited by 0
1 more
  1. 2026Fully Kolmogorov-Arnold Deep Model in Medical Image Segmentationcited by 0