FCKAN: Evaluating KAN for Time Series Classification and Extrinsic Regression

• Gabriel da Costa Merlin USP
• Adilson Junior Alves Medronha USP
• Diego Furtado Silva USP

DOI: https://doi.org/10.5753/eniac.2025.13884

Abstract

Time Series Classification (TSC) and Time Series Extrinsic Regression (TSER) are critical tasks across diverse fields. While Fully Convolutional Networks (FCNs) effectively capture temporal dependencies, Kolmogorov–Arnold Networks (KANs) offer greater flexibility and interpretability. However, integrating KANs with temporal encoders and their application to regression tasks remain largely unexplored. This paper introduces FCKAN and Hybrid FCN-KAN, two novel architectures that combine FCNs and KANs for TSC and TSER. The first is an end-to-end model, while the second is a hybrid approach that leverages a pre-trained FCN as a feature extractor followed by a KAN. We conduct experiments on 147 benchmark datasets. For TSC, both architectures outperform non-temporal baselines and achieve competitive performance with FCNs. In TSER, although all models are statistically equivalent, temporal models consistently outperform non-temporal baselines.

References

Ang, L.-M. and Seng, K. P. (2016). Big sensor data applications in urban environments. Big Data Research, 4:1–12.

Bagnall, A., Bostrom, A., Large, J., and Lines, J. (2016). The great time series classification bake off: An experimental evaluation of recently proposed algorithms. extended version.

Dau, H. A., Keogh, E., Kamgar, K., Yeh, C.-C. M., Zhu, Y., Gharghabi, S., Ratanamahatana, C. A., Yanping, Hu, B., Begum, N., Bagnall, A., Mueen, A., and Batista, G. (2018). The ucr time series classification archive. [link].

Dong, C., Zheng, L., and Chen, W. (2024). Kolmogorov-arnold networks (kan) for time series classification and robust analysis.

El Maachi, I., Bilodeau, G.-A., and Bouachir, W. (2020). Deep 1d-convnet for accurate parkinson disease detection and severity prediction from gait. Expert Systems with Applications, 143:113075.

Ismail-Fawaz, A., Devanne, M., Weber, J., and Forestier, G. (2022). Deep learning for time series classification using new hand-crafted convolution filters. pages 972–981.

Ismail Fawaz, H., Forestier, G., Weber, J., Idoumghar, L., and Muller, P.-A. (2019). Deep learning for time series classification: a review. Data Mining and Knowledge Discovery, 33(4):917–963.

Ismail Fawaz, H., Lucas, B., Forestier, G., Pelletier, C., Schmidt, D. F., Weber, J., Webb, G. I., Idoumghar, L., Muller, P.-A., and Petitjean, F. (2020). Inceptiontime: Finding alexnet for time series classification. Data Mining and Knowledge Discovery, 34(6):1936–1962.

Lim, B. and Zohren, S. (2021). Time-series forecasting with deep learning: a survey. Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, 379(2194):20200209.

Liu, Z., Wang, Y., Vaidya, S., Ruehle, F., Halverson, J., Soljačić, M., Hou, T. Y., and Tegmark, M. (2024). Kan: Kolmogorov-arnold networks.

Middlehurst, M., Ismail-Fawaz, A., Guillaume, A., Holder, C., Guijo-Rubio, D., Bulatova, G., Tsaprounis, L., Mentel, L., Walter, M., Schäfer, P., and Bagnall, A. (2024). aeon: a python toolkit for learning from time series. Journal of Machine Learning Research, 25(289):1–10.

Paszke, A., Gross, S., Massa, F., Lerer, A., Bradbury, J., Chanan, G., Killeen, T., Lin, Z., Gimelshein, N., Antiga, L., Desmaison, A., Kopf, A., Yang, E., DeVito, Z., Raison, M., Tejani, A., Chilamkurthy, S., Steiner, B., Fang, L., Bai, J., and Chintala, S. (2019). Pytorch: An imperative style, high-performance deep learning library. In Advances in Neural Information Processing Systems 32, pages 8024–8035. Curran Associates, Inc.

Tan, C. W., Bergmeir, C., Petitjean, F., and Webb, G. I. (2021a). Time series extrinsic regression. Data Mining and Knowledge Discovery, pages 1–29.

Tan, C. W., Bergmeir, C., Petitjean, F., and Webb, G. I. (2021b). Time series extrinsic regression. Data Mining and Knowledge Discovery, 35(3):1032–1060.

Wang, Z., Yan, W., and Oates, T. (2016). Time series classification from scratch with deep neural networks: A strong baseline.