Regularizing Temporal Explanations in Dynamic Neural Networks
Electronics · 20.05.2026 · 15 (10) · 2200
DOI: 10.3390/electronics15102200 ↗Kurz gefasst
Fachartikel in der Zeitschrift Electronics, online erschienen am 20.05.2026. Verfasst von Dalius Navakauskas und 1 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 33 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Using attribution-based priors to improve the temporal interpretability and robustness of dynamic neural networks provides a computationally efficient method that does not alter the model structure during inference. We explore explanation-guided training for timeseries classification through the introduction of attribution-sensitive loss terms that serve as regularizers for the evolution of input relevance over time. The main contributions are the Temporal Relevance Smoothness Index (TRSI) and a ratio-based loss that reduces irregular step-to-step changes in channel-aggregated absolute relevance. TRSI is compared against temporal total-variation penalties computed using Layer-wise Relevance Propagation Total Variation (LRP-TV) and Integrated Gradients Total Variation (IG-TV). Experiments on a controlled three-class subset of the Korean University Human Activity Recognition (KU-HAR) dataset using a finite impulse response neural network (FIRNN) show that TRSI yields the strongest smoothness improvement, reducing the total variation of the aggregated relevance signal from 0.768 to 0.447 (41.8%), compared with 0.667 (LRP-TV) and 0.677 (IG-TV). Robustness tests indicate a clear advantage for TRSI under impulsive and white Gaussian test-time noise.
Zitieren
APA
Navakauskas, D., & Dumpis, M. (2026). Regularizing Temporal Explanations in Dynamic Neural Networks. Electronics, 15(10), 2200. https://doi.org/10.3390/electronics15102200
Vancouver
Navakauskas D, Dumpis M. Regularizing Temporal Explanations in Dynamic Neural Networks. Electronics. 2026;15(10):2200. doi:10.3390/electronics15102200
BibTeX
@article{Navakauskas2026,
author = {Navakauskas, Dalius and Dumpis, Martynas},
title = {Regularizing Temporal Explanations in Dynamic Neural Networks},
journal = {Electronics},
year = {2026},
volume = {15},
number = {10},
pages = {2200},
doi = {10.3390/electronics15102200}
}
EUVIMED