Efficient Transformations in Deep Learning Convolutional Neural Networks

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Yilmaz, Berk, Harvey, Daniel Fidel, Dhuri, Prajit
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915352318836736
author Yilmaz, Berk
Harvey, Daniel Fidel
Dhuri, Prajit
author_facet Yilmaz, Berk
Harvey, Daniel Fidel
Dhuri, Prajit
contents This study investigates the integration of signal processing transformations -- Fast Fourier Transform (FFT), Walsh-Hadamard Transform (WHT), and Discrete Cosine Transform (DCT) -- within the ResNet50 convolutional neural network (CNN) model for image classification. The primary objective is to assess the trade-offs between computational efficiency, energy consumption, and classification accuracy during training and inference. Using the CIFAR-100 dataset (100 classes, 60,000 images), experiments demonstrated that incorporating WHT significantly reduced energy consumption while improving accuracy. Specifically, a baseline ResNet50 model achieved a testing accuracy of 66%, consuming an average of 25,606 kJ per model. In contrast, a modified ResNet50 incorporating WHT in the early convolutional layers achieved 74% accuracy, and an enhanced version with WHT applied to both early and late layers achieved 79% accuracy, with an average energy consumption of only 39 kJ per model. These results demonstrate the potential of WHT as a highly efficient and effective approach for energy-constrained CNN applications.
format Preprint
id arxiv_https___arxiv_org_abs_2506_16418
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Efficient Transformations in Deep Learning Convolutional Neural Networks
Yilmaz, Berk
Harvey, Daniel Fidel
Dhuri, Prajit
Computer Vision and Pattern Recognition
Artificial Intelligence
Image and Video Processing
Signal Processing
68T07, 68T10, 94A08, 42C10
This study investigates the integration of signal processing transformations -- Fast Fourier Transform (FFT), Walsh-Hadamard Transform (WHT), and Discrete Cosine Transform (DCT) -- within the ResNet50 convolutional neural network (CNN) model for image classification. The primary objective is to assess the trade-offs between computational efficiency, energy consumption, and classification accuracy during training and inference. Using the CIFAR-100 dataset (100 classes, 60,000 images), experiments demonstrated that incorporating WHT significantly reduced energy consumption while improving accuracy. Specifically, a baseline ResNet50 model achieved a testing accuracy of 66%, consuming an average of 25,606 kJ per model. In contrast, a modified ResNet50 incorporating WHT in the early convolutional layers achieved 74% accuracy, and an enhanced version with WHT applied to both early and late layers achieved 79% accuracy, with an average energy consumption of only 39 kJ per model. These results demonstrate the potential of WHT as a highly efficient and effective approach for energy-constrained CNN applications.
title Efficient Transformations in Deep Learning Convolutional Neural Networks
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Image and Video Processing
Signal Processing
68T07, 68T10, 94A08, 42C10
url https://arxiv.org/abs/2506.16418