Cover Image

Identification of untrained class data using neuron clusters

CC BY

Saved in:
Bibliographic Details
Main Authors: Young-Woo, Lee, Heung-Seok, Chae
Format: Book
Language:English
Published: Springer 2023
Subjects:
CNNs
Online Access:https://link.springer.com/article/10.1007/s00521-023-08265-x
https://dlib.phenikaa-uni.edu.vn/handle/PNK/8297
Tags: Add Tag
No Tags, Be the first to tag this record!
id oai:localhost:PNK-8297
record_format dspace
spelling oai:localhost:PNK-82972023-04-25T08:13:06Z Identification of untrained class data using neuron clusters Young-Woo, Lee Heung-Seok, Chae CNNs CC BY Convolutional neural networks (CNNs), a representative type of deep neural networks, are used in various fields. There are problems that should be solved to operate CNN in the real-world. In real-world operating environments, the CNN’s performance may be degraded due to data of untrained types, which limits its operability. In this study, we propose a method for identifying data of a type that the model has not trained on based on the neuron cluster, a set of neurons activated based on the type of input data. In experiments performed on the ResNet model with the MNIST, CIFAR-10, and STL-10 datasets, the proposed method identifies data of untrained and trained types with an accuracy of 85% or higher. The more data used for neuron cluster identification, the higher the accuracy; conversely, the more complex the dataset's characteristics, the lower the accuracy. The proposed method uses only the information of activated neurons without any addition or modification of the model’s structure; hence, the computational cost is low without affecting the classification performance of the model. 2023-04-25T08:13:06Z 2023-04-25T08:13:06Z 2023 Book https://link.springer.com/article/10.1007/s00521-023-08265-x https://dlib.phenikaa-uni.edu.vn/handle/PNK/8297 en application/pdf Springer
institution Digital Phenikaa
collection Digital Phenikaa
language English
topic CNNs
spellingShingle CNNs
Young-Woo, Lee
Heung-Seok, Chae
Identification of untrained class data using neuron clusters
description CC BY
format Book
author Young-Woo, Lee
Heung-Seok, Chae
author_facet Young-Woo, Lee
Heung-Seok, Chae
author_sort Young-Woo, Lee
title Identification of untrained class data using neuron clusters
title_short Identification of untrained class data using neuron clusters
title_full Identification of untrained class data using neuron clusters
title_fullStr Identification of untrained class data using neuron clusters
title_full_unstemmed Identification of untrained class data using neuron clusters
title_sort identification of untrained class data using neuron clusters
publisher Springer
publishDate 2023
url https://link.springer.com/article/10.1007/s00521-023-08265-x
https://dlib.phenikaa-uni.edu.vn/handle/PNK/8297
_version_ 1764177439135629312
score 8.891145

Similar Items