Efficient Neural Networks with Discrete Cosine Transform Activations

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Martinez-Gost, Marc, Pepe, Sara, Pérez-Neira, Ana, Lagunas, Miguel Ángel
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918187340136448
author Martinez-Gost, Marc
Pepe, Sara
Pérez-Neira, Ana
Lagunas, Miguel Ángel
author_facet Martinez-Gost, Marc
Pepe, Sara
Pérez-Neira, Ana
Lagunas, Miguel Ángel
contents In this paper, we extend our previous work on the Expressive Neural Network (ENN), a multilayer perceptron with adaptive activation functions parametrized using the Discrete Cosine Transform (DCT). Building upon previous work that demonstrated the strong expressiveness of ENNs with compact architectures, we now emphasize their efficiency, interpretability and pruning capabilities. The DCT-based parameterization provides a structured and decorrelated representation that reveals the functional role of each neuron and allows direct identification of redundant components. Leveraging this property, we propose an efficient pruning strategy that removes unnecessary DCT coefficients with negligible or no loss in performance. Experimental results across classification and implicit neural representation tasks confirm that ENNs achieve state-of-the-art accuracy while maintaining a low number of parameters. Furthermore, up to 40% of the activation coefficients can be safely pruned, thanks to the orthogonality and bounded nature of the DCT basis. Overall, these findings demonstrate that the ENN framework offers a principled integration of signal processing concepts into neural network design, achieving a balanced trade-off between expressiveness, compactness, and interpretability.
format Preprint
id arxiv_https___arxiv_org_abs_2511_03531
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Efficient Neural Networks with Discrete Cosine Transform Activations
Martinez-Gost, Marc
Pepe, Sara
Pérez-Neira, Ana
Lagunas, Miguel Ángel
Machine Learning
Artificial Intelligence
In this paper, we extend our previous work on the Expressive Neural Network (ENN), a multilayer perceptron with adaptive activation functions parametrized using the Discrete Cosine Transform (DCT). Building upon previous work that demonstrated the strong expressiveness of ENNs with compact architectures, we now emphasize their efficiency, interpretability and pruning capabilities. The DCT-based parameterization provides a structured and decorrelated representation that reveals the functional role of each neuron and allows direct identification of redundant components. Leveraging this property, we propose an efficient pruning strategy that removes unnecessary DCT coefficients with negligible or no loss in performance. Experimental results across classification and implicit neural representation tasks confirm that ENNs achieve state-of-the-art accuracy while maintaining a low number of parameters. Furthermore, up to 40% of the activation coefficients can be safely pruned, thanks to the orthogonality and bounded nature of the DCT basis. Overall, these findings demonstrate that the ENN framework offers a principled integration of signal processing concepts into neural network design, achieving a balanced trade-off between expressiveness, compactness, and interpretability.
title Efficient Neural Networks with Discrete Cosine Transform Activations
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2511.03531