EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Rethinking the Role of the Classification Head in CNN-BiLSTM-Based Human Activity Recognition

Junbing Shi, Yingmin Wang, Libo Yang

Electronics · 09.11.2026 · 15 (21) · 5194

DOI: 10.3390/electronics15215194 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Electronics, online erschienen am 09.11.2026. Verfasst von Junbing Shi und 2 weiteren Autor:innen. Die Arbeit nennt 52 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Shanxi Higher Education Teaching Reform and Innovation Project.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Wearable sensor-based human activity recognition (HAR) underpins healthcare monitoring, fall detection, and fitness tracking. HAR pipelines pair a feature-extraction backbone with a classification head. Backbone design is well studied, whereas the classification head has received less attention. We ask how much the choice of the classification head matters. Holding the CNN-BiLSTM backbone architecture fixed across seven public HAR datasets under leave-one-subject-out cross-validation, we compare four heads: a 774-parameter linear projection, a small multi-layer perceptron (MLP), a Matched MLP, and a parameter-matched Gaussian radial-basis-function Kolmogorov–Arnold network (KAN). The classification head has a limited effect. The best-to-worst spread across heads is under 3 percentage points (pp), and the 774-parameter linear head stays within 1.5 pp of the best head on every dataset. When the backbone is frozen and each head is retrained on the fixed features, the spread narrows to about 1 pp on six of seven datasets, indicating that additional head capacity has little effect once the representation is fixed. Under parameter matching, the KAN head ranks below the Matched MLP on six of seven datasets but first on HHAR, with effect sizes near zero in every KAN–MLP comparison. A minimal linear head therefore matches the accuracy of more complex heads at lower full-model latency than the KAN head on both GPU and CPU (2.10 versus 3.90 ms and 4.37 versus 5.55 ms), and at 0.5% of the KAN head’s parameter count.

Förderung

  • Shanxi Higher Education Teaching Reform and Innovation Project

Zitieren

APA

Shi, J., Wang, Y., & Yang, L. (2026). Rethinking the Role of the Classification Head in CNN-BiLSTM-Based Human Activity Recognition. Electronics, 15(21), 5194. https://doi.org/10.3390/electronics15215194

Vancouver

Shi J, Wang Y, Yang L. Rethinking the Role of the Classification Head in CNN-BiLSTM-Based Human Activity Recognition. Electronics. 2026;15(21):5194. doi:10.3390/electronics15215194

BibTeX
@article{Shi2026,
  author = {Shi, Junbing and Wang, Yingmin and Yang, Libo},
  title = {Rethinking the Role of the Classification Head in CNN-BiLSTM-Based Human Activity Recognition},
  journal = {Electronics},
  year = {2026},
  volume = {15},
  number = {21},
  pages = {5194},
  doi = {10.3390/electronics15215194}
}
Kostenlos starten →