Enhancing Task-Incremental Learning via a Prompt-Based Hybrid Convolutional Neural Networks (CNNs)-Vision Transformer (ViT) Framework

Artificial neural network (ANN) models are widely used in various fields such as image classification, multi-object detection, intent prediction, military applications, and natural language processing. However, artificial intelligence (AI) models for continual learning (CL) are not yet mature, and &...

Full description

Saved in:
Bibliographic Details
Main Authors: Zuomin Yang, Anis Salwa Mohd Khairuddin, Joon Huang Chuah, Wei Ru Wong, Xin Xu, Hafiz Muhammad Fahad Noman, Qiyuan Qin
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/11121184/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849223088057614336
author Zuomin Yang
Anis Salwa Mohd Khairuddin
Joon Huang Chuah
Wei Ru Wong
Xin Xu
Hafiz Muhammad Fahad Noman
Qiyuan Qin
author_facet Zuomin Yang
Anis Salwa Mohd Khairuddin
Joon Huang Chuah
Wei Ru Wong
Xin Xu
Hafiz Muhammad Fahad Noman
Qiyuan Qin
author_sort Zuomin Yang
collection DOAJ
description Artificial neural network (ANN) models are widely used in various fields such as image classification, multi-object detection, intent prediction, military applications, and natural language processing. However, artificial intelligence (AI) models for continual learning (CL) are not yet mature, and “catastrophic forgetting (CF)” is still a typical problem. The study of biological neural networks (BNNs) and ANN models still needs further exploration. Therefore, this paper mainly explores the pre- and postsynaptic structures, the synaptic cleft, the early and late stages of long-term potentiation, and the effects of neurotransmitters on synaptic excitation and inhibition. We emphasize the necessity of integrating biological neural systems and ANN models in learning and memory. Based on the “Prompt Pool”, this paper designs a hybrid neural network (HNN) architecture that integrates convolutional neural networks (CNNs), vision transformers (ViT), prompt pools, and adapters to alleviate the “CF” problem in task incremental learning (TIL). Compared with the existing ViT and prompt pool architecture, this method shows higher performance in final task training and also shows certain advantages in the persistence of TIL. In the future, based on the principles of biological neuroscience, we will further apply the HNN model to image classification and multi-object detection tasks in autonomous driving. By gaining a deeper understanding of the BNN mechanisms, we will develop efficient HNN models that can adapt to dynamic environments and provide new solutions for CL.
format Article
id doaj-art-2f267e4a57e6426d9a20f3fa5d5aba87
institution Kabale University
issn 2169-3536
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-2f267e4a57e6426d9a20f3fa5d5aba872025-08-25T23:11:56ZengIEEEIEEE Access2169-35362025-01-011314522314524210.1109/ACCESS.2025.359702011121184Enhancing Task-Incremental Learning via a Prompt-Based Hybrid Convolutional Neural Networks (CNNs)-Vision Transformer (ViT) FrameworkZuomin Yang0https://orcid.org/0000-0003-4776-4887Anis Salwa Mohd Khairuddin1https://orcid.org/0000-0002-9873-4779Joon Huang Chuah2https://orcid.org/0000-0001-9058-3497Wei Ru Wong3https://orcid.org/0000-0002-7643-6172Xin Xu4https://orcid.org/0000-0002-9194-1249Hafiz Muhammad Fahad Noman5https://orcid.org/0000-0001-8507-5383Qiyuan Qin6https://orcid.org/0009-0005-2736-7667Department of Electrical Engineering, Faculty of Engineering, Universiti Malaya, Kuala Lumpur, MalaysiaDepartment of Electrical Engineering, Faculty of Engineering, Universiti Malaya, Kuala Lumpur, MalaysiaDepartment of Electrical Engineering, Faculty of Engineering, Universiti Malaya, Kuala Lumpur, MalaysiaDepartment of Electrical Engineering, Faculty of Engineering, Universiti Malaya, Kuala Lumpur, MalaysiaDepartment of Mechanical Engineering, Faculty of Engineering, Universiti Malaya, Kuala Lumpur, MalaysiaDepartment of Electrical Engineering, Faculty of Engineering, Universiti Malaya, Kuala Lumpur, MalaysiaDepartment of Electrical Engineering, Faculty of Engineering, Universiti Malaya, Kuala Lumpur, MalaysiaArtificial neural network (ANN) models are widely used in various fields such as image classification, multi-object detection, intent prediction, military applications, and natural language processing. However, artificial intelligence (AI) models for continual learning (CL) are not yet mature, and “catastrophic forgetting (CF)” is still a typical problem. The study of biological neural networks (BNNs) and ANN models still needs further exploration. Therefore, this paper mainly explores the pre- and postsynaptic structures, the synaptic cleft, the early and late stages of long-term potentiation, and the effects of neurotransmitters on synaptic excitation and inhibition. We emphasize the necessity of integrating biological neural systems and ANN models in learning and memory. Based on the “Prompt Pool”, this paper designs a hybrid neural network (HNN) architecture that integrates convolutional neural networks (CNNs), vision transformers (ViT), prompt pools, and adapters to alleviate the “CF” problem in task incremental learning (TIL). Compared with the existing ViT and prompt pool architecture, this method shows higher performance in final task training and also shows certain advantages in the persistence of TIL. In the future, based on the principles of biological neuroscience, we will further apply the HNN model to image classification and multi-object detection tasks in autonomous driving. By gaining a deeper understanding of the BNN mechanisms, we will develop efficient HNN models that can adapt to dynamic environments and provide new solutions for CL.https://ieeexplore.ieee.org/document/11121184/Artificial neural networkscontinual learninghybrid neural networkssynaptic plasticityvision transformer
spellingShingle Zuomin Yang
Anis Salwa Mohd Khairuddin
Joon Huang Chuah
Wei Ru Wong
Xin Xu
Hafiz Muhammad Fahad Noman
Qiyuan Qin
Enhancing Task-Incremental Learning via a Prompt-Based Hybrid Convolutional Neural Networks (CNNs)-Vision Transformer (ViT) Framework
IEEE Access
Artificial neural networks
continual learning
hybrid neural networks
synaptic plasticity
vision transformer
title Enhancing Task-Incremental Learning via a Prompt-Based Hybrid Convolutional Neural Networks (CNNs)-Vision Transformer (ViT) Framework
title_full Enhancing Task-Incremental Learning via a Prompt-Based Hybrid Convolutional Neural Networks (CNNs)-Vision Transformer (ViT) Framework
title_fullStr Enhancing Task-Incremental Learning via a Prompt-Based Hybrid Convolutional Neural Networks (CNNs)-Vision Transformer (ViT) Framework
title_full_unstemmed Enhancing Task-Incremental Learning via a Prompt-Based Hybrid Convolutional Neural Networks (CNNs)-Vision Transformer (ViT) Framework
title_short Enhancing Task-Incremental Learning via a Prompt-Based Hybrid Convolutional Neural Networks (CNNs)-Vision Transformer (ViT) Framework
title_sort enhancing task incremental learning via a prompt based hybrid convolutional neural networks cnns vision transformer vit framework
topic Artificial neural networks
continual learning
hybrid neural networks
synaptic plasticity
vision transformer
url https://ieeexplore.ieee.org/document/11121184/
work_keys_str_mv AT zuominyang enhancingtaskincrementallearningviaapromptbasedhybridconvolutionalneuralnetworkscnnsvisiontransformervitframework
AT anissalwamohdkhairuddin enhancingtaskincrementallearningviaapromptbasedhybridconvolutionalneuralnetworkscnnsvisiontransformervitframework
AT joonhuangchuah enhancingtaskincrementallearningviaapromptbasedhybridconvolutionalneuralnetworkscnnsvisiontransformervitframework
AT weiruwong enhancingtaskincrementallearningviaapromptbasedhybridconvolutionalneuralnetworkscnnsvisiontransformervitframework
AT xinxu enhancingtaskincrementallearningviaapromptbasedhybridconvolutionalneuralnetworkscnnsvisiontransformervitframework
AT hafizmuhammadfahadnoman enhancingtaskincrementallearningviaapromptbasedhybridconvolutionalneuralnetworkscnnsvisiontransformervitframework
AT qiyuanqin enhancingtaskincrementallearningviaapromptbasedhybridconvolutionalneuralnetworkscnnsvisiontransformervitframework