Adaptive Threshold Hierarchical Incremental Learning Method
Traditional deep convolutional neural networks have achieved excellent performance on various machine learning tasks. Still, they perform poorly in continuous data stream environments, where models trained on new datasets often suffer from a significant drop in performance on old datasets, a phenome...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
IEEE
2023-01-01
|
Series: | IEEE Access |
Subjects: | |
Online Access: | https://ieeexplore.ieee.org/document/10038576/ |
_version_ | 1811163518176591872 |
---|---|
author | Xingyu Li Shengbo Dong Qiya Su Muyao Yu Xinzhi Li |
author_facet | Xingyu Li Shengbo Dong Qiya Su Muyao Yu Xinzhi Li |
author_sort | Xingyu Li |
collection | DOAJ |
description | Traditional deep convolutional neural networks have achieved excellent performance on various machine learning tasks. Still, they perform poorly in continuous data stream environments, where models trained on new datasets often suffer from a significant drop in performance on old datasets, a phenomenon known as “catastrophic forgetting.” Incremental learning can help solve the “catastrophic forgetting” problem in deep learning by learning new knowledge while retaining what has already been learned. In practice, incremental learning algorithms usually need to be deployed on edge devices with limited memory and restricted access to training data, facing the problems of high model complexity and imbalance between old and new categories of data. We propose an Adaptive Threshold Hierarchical Incremental Learning (ATHIL) method to address the above problems. Our proposed method does not require additional data and model storage space during the training process, combines local weight discrete coefficient thresholding and the mean nearest neighbor principle, uses a sparse matrix hierarchical masking network, and flexibly adjusts the network structure according to different tasks to achieve learning multiple image classification tasks in a single network. The experimental results show that the performance of the proposed method significantly outperforms existing methods on fine-grained classification datasets under three evaluation metrics. |
first_indexed | 2024-04-10T15:07:35Z |
format | Article |
id | doaj.art-8416e5e5a7004c58b1ac3387d07e2512 |
institution | Directory Open Access Journal |
issn | 2169-3536 |
language | English |
last_indexed | 2024-04-10T15:07:35Z |
publishDate | 2023-01-01 |
publisher | IEEE |
record_format | Article |
series | IEEE Access |
spelling | doaj.art-8416e5e5a7004c58b1ac3387d07e25122023-02-15T00:00:16ZengIEEEIEEE Access2169-35362023-01-0111122851229310.1109/ACCESS.2023.324268810038576Adaptive Threshold Hierarchical Incremental Learning MethodXingyu Li0https://orcid.org/0000-0002-8464-0106Shengbo Dong1Qiya Su2Muyao Yu3Xinzhi Li4Beijing Institute of Remote Sensing Equipment, Beijing, ChinaBeijing Institute of Remote Sensing Equipment, Beijing, ChinaBeijing Institute of Remote Sensing Equipment, Beijing, ChinaBeijing Institute of Remote Sensing Equipment, Beijing, ChinaBeijing Institute of Remote Sensing Equipment, Beijing, ChinaTraditional deep convolutional neural networks have achieved excellent performance on various machine learning tasks. Still, they perform poorly in continuous data stream environments, where models trained on new datasets often suffer from a significant drop in performance on old datasets, a phenomenon known as “catastrophic forgetting.” Incremental learning can help solve the “catastrophic forgetting” problem in deep learning by learning new knowledge while retaining what has already been learned. In practice, incremental learning algorithms usually need to be deployed on edge devices with limited memory and restricted access to training data, facing the problems of high model complexity and imbalance between old and new categories of data. We propose an Adaptive Threshold Hierarchical Incremental Learning (ATHIL) method to address the above problems. Our proposed method does not require additional data and model storage space during the training process, combines local weight discrete coefficient thresholding and the mean nearest neighbor principle, uses a sparse matrix hierarchical masking network, and flexibly adjusts the network structure according to different tasks to achieve learning multiple image classification tasks in a single network. The experimental results show that the performance of the proposed method significantly outperforms existing methods on fine-grained classification datasets under three evaluation metrics.https://ieeexplore.ieee.org/document/10038576/Incremental learningconvolutional neural networkdeep learning |
spellingShingle | Xingyu Li Shengbo Dong Qiya Su Muyao Yu Xinzhi Li Adaptive Threshold Hierarchical Incremental Learning Method IEEE Access Incremental learning convolutional neural network deep learning |
title | Adaptive Threshold Hierarchical Incremental Learning Method |
title_full | Adaptive Threshold Hierarchical Incremental Learning Method |
title_fullStr | Adaptive Threshold Hierarchical Incremental Learning Method |
title_full_unstemmed | Adaptive Threshold Hierarchical Incremental Learning Method |
title_short | Adaptive Threshold Hierarchical Incremental Learning Method |
title_sort | adaptive threshold hierarchical incremental learning method |
topic | Incremental learning convolutional neural network deep learning |
url | https://ieeexplore.ieee.org/document/10038576/ |
work_keys_str_mv | AT xingyuli adaptivethresholdhierarchicalincrementallearningmethod AT shengbodong adaptivethresholdhierarchicalincrementallearningmethod AT qiyasu adaptivethresholdhierarchicalincrementallearningmethod AT muyaoyu adaptivethresholdhierarchicalincrementallearningmethod AT xinzhili adaptivethresholdhierarchicalincrementallearningmethod |