Deep Multi-Component Neural Network Architecture

Existing neural network architectures often struggle with two critical limitations: (1) information loss during dataset length standardization, where variable-length samples are forced into fixed dimensions, and (2) inefficient feature selection in single-modal systems, which treats all features equ...

Full description

Saved in:
Bibliographic Details
Main Authors: Chafik Boulealam, Hajar Filali, Jamal Riffi, Adnane Mohamed Mahraz, Hamid Tairi
Format: Article
Language:English
Published: MDPI AG 2025-04-01
Series:Computation
Subjects:
Online Access:https://www.mdpi.com/2079-3197/13/4/93
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850183758765883392
author Chafik Boulealam
Hajar Filali
Jamal Riffi
Adnane Mohamed Mahraz
Hamid Tairi
author_facet Chafik Boulealam
Hajar Filali
Jamal Riffi
Adnane Mohamed Mahraz
Hamid Tairi
author_sort Chafik Boulealam
collection DOAJ
description Existing neural network architectures often struggle with two critical limitations: (1) information loss during dataset length standardization, where variable-length samples are forced into fixed dimensions, and (2) inefficient feature selection in single-modal systems, which treats all features equally regardless of relevance. To address these issues, this paper introduces the Deep Multi-Components Neural Network (DMCNN), a novel architecture that processes variable-length data by regrouping samples into components of similar lengths, thereby preserving information that traditional methods discard. DMCNN dynamically prioritizes task-relevant features through a component-weighting mechanism, which calculates the importance of each component via loss functions and adjusts weights using a SoftMax function. This approach eliminates the need for dataset standardization while enhancing meaningful features and suppressing irrelevant ones. Additionally, DMCNN seamlessly integrates multimodal data (e.g., text, speech, and signals) as separate components, leveraging complementary information to improve accuracy without requiring dimension alignment. Evaluated on the Multimodal EmotionLines Dataset (MELD) and CIFAR-10, DMCNN achieves state-of-the-art accuracy of 99.22% on MELD and 97.78% on CIFAR-10, outperforming existing methods like MNN and McDFR. The architecture’s efficiency is further demonstrated by its reduced trainable parameters and robust handling of multimodal and variable-length inputs, making it a versatile solution for classification tasks.
format Article
id doaj-art-fffa9f773a4d47759310786b1179d98a
institution OA Journals
issn 2079-3197
language English
publishDate 2025-04-01
publisher MDPI AG
record_format Article
series Computation
spelling doaj-art-fffa9f773a4d47759310786b1179d98a2025-08-20T02:17:14ZengMDPI AGComputation2079-31972025-04-011349310.3390/computation13040093Deep Multi-Component Neural Network ArchitectureChafik Boulealam0Hajar Filali1Jamal Riffi2Adnane Mohamed Mahraz3Hamid Tairi4LISAC, Department of Computer Science, Faculty of Science Dhar El Mahraz, Sidi Mohamed Ben Abdellah University, Fez 30000, MoroccoLISAC, Department of Computer Science, Faculty of Science Dhar El Mahraz, Sidi Mohamed Ben Abdellah University, Fez 30000, MoroccoLISAC, Department of Computer Science, Faculty of Science Dhar El Mahraz, Sidi Mohamed Ben Abdellah University, Fez 30000, MoroccoLISAC, Department of Computer Science, Faculty of Science Dhar El Mahraz, Sidi Mohamed Ben Abdellah University, Fez 30000, MoroccoLISAC, Department of Computer Science, Faculty of Science Dhar El Mahraz, Sidi Mohamed Ben Abdellah University, Fez 30000, MoroccoExisting neural network architectures often struggle with two critical limitations: (1) information loss during dataset length standardization, where variable-length samples are forced into fixed dimensions, and (2) inefficient feature selection in single-modal systems, which treats all features equally regardless of relevance. To address these issues, this paper introduces the Deep Multi-Components Neural Network (DMCNN), a novel architecture that processes variable-length data by regrouping samples into components of similar lengths, thereby preserving information that traditional methods discard. DMCNN dynamically prioritizes task-relevant features through a component-weighting mechanism, which calculates the importance of each component via loss functions and adjusts weights using a SoftMax function. This approach eliminates the need for dataset standardization while enhancing meaningful features and suppressing irrelevant ones. Additionally, DMCNN seamlessly integrates multimodal data (e.g., text, speech, and signals) as separate components, leveraging complementary information to improve accuracy without requiring dimension alignment. Evaluated on the Multimodal EmotionLines Dataset (MELD) and CIFAR-10, DMCNN achieves state-of-the-art accuracy of 99.22% on MELD and 97.78% on CIFAR-10, outperforming existing methods like MNN and McDFR. The architecture’s efficiency is further demonstrated by its reduced trainable parameters and robust handling of multimodal and variable-length inputs, making it a versatile solution for classification tasks.https://www.mdpi.com/2079-3197/13/4/93deep learning (DL)image classificationmeaningful neural network (MNN)multimodaldeep multi-components neural network architecture (DMCNN)neural network architecture
spellingShingle Chafik Boulealam
Hajar Filali
Jamal Riffi
Adnane Mohamed Mahraz
Hamid Tairi
Deep Multi-Component Neural Network Architecture
Computation
deep learning (DL)
image classification
meaningful neural network (MNN)
multimodal
deep multi-components neural network architecture (DMCNN)
neural network architecture
title Deep Multi-Component Neural Network Architecture
title_full Deep Multi-Component Neural Network Architecture
title_fullStr Deep Multi-Component Neural Network Architecture
title_full_unstemmed Deep Multi-Component Neural Network Architecture
title_short Deep Multi-Component Neural Network Architecture
title_sort deep multi component neural network architecture
topic deep learning (DL)
image classification
meaningful neural network (MNN)
multimodal
deep multi-components neural network architecture (DMCNN)
neural network architecture
url https://www.mdpi.com/2079-3197/13/4/93
work_keys_str_mv AT chafikboulealam deepmulticomponentneuralnetworkarchitecture
AT hajarfilali deepmulticomponentneuralnetworkarchitecture
AT jamalriffi deepmulticomponentneuralnetworkarchitecture
AT adnanemohamedmahraz deepmulticomponentneuralnetworkarchitecture
AT hamidtairi deepmulticomponentneuralnetworkarchitecture