2N labeling defense method against adversarial attacks by filtering and extended class label set
CC BY
Lưu vào:
Tác giả chính: | , |
---|---|
Định dạng: | Sách |
Ngôn ngữ: | English |
Nhà xuất bản: |
Springer
2023
|
Chủ đề: | |
Truy cập trực tuyến: | https://link.springer.com/article/10.1007/s11042-022-14021-5 https://dlib.phenikaa-uni.edu.vn/handle/PNK/8333 |
Từ khóa: |
Thêm từ khóa
Không có từ khóa, Hãy là người đầu tiên đánh dấu biểu ghi này!
|
id |
oai:localhost:PNK-8333 |
---|---|
record_format |
dspace |
spelling |
oai:localhost:PNK-83332023-04-26T06:44:51Z 2N labeling defense method against adversarial attacks by filtering and extended class label set Gábor, Szűcs Richárd, Kiss 2N labeling defense method NULL labeling method CC BY The fast improvement of deep learning methods resulted in breakthroughs in image classification, however, these models are sensitive to adversarial perturbations, which can cause serious problems. Adversarial attacks try to change the model output by adding noise to the input, in our research we propose a combined defense method against it. Two defense approaches have been evolved in the literature, one robustizes the attacked model for higher accuracy, and the other approach detects the adversarial examples. Only very few papers discuss both approaches, thus our aim was to combine them to obtain a more robust model and to examine the combination, in particular the filtering capability of the detector. Our contribution was that the filtering based on the decision of the detector is able to enhance the accuracy, which was theoretically proved. 2023-04-26T06:44:51Z 2023-04-26T06:44:51Z 2022 Book https://link.springer.com/article/10.1007/s11042-022-14021-5 https://dlib.phenikaa-uni.edu.vn/handle/PNK/8333 en application/pdf Springer |
institution |
Digital Phenikaa |
collection |
Digital Phenikaa |
language |
English |
topic |
2N labeling defense method NULL labeling method |
spellingShingle |
2N labeling defense method NULL labeling method Gábor, Szűcs Richárd, Kiss 2N labeling defense method against adversarial attacks by filtering and extended class label set |
description |
CC BY |
format |
Book |
author |
Gábor, Szűcs Richárd, Kiss |
author_facet |
Gábor, Szűcs Richárd, Kiss |
author_sort |
Gábor, Szűcs |
title |
2N labeling defense method against adversarial attacks by filtering and extended class label set |
title_short |
2N labeling defense method against adversarial attacks by filtering and extended class label set |
title_full |
2N labeling defense method against adversarial attacks by filtering and extended class label set |
title_fullStr |
2N labeling defense method against adversarial attacks by filtering and extended class label set |
title_full_unstemmed |
2N labeling defense method against adversarial attacks by filtering and extended class label set |
title_sort |
2n labeling defense method against adversarial attacks by filtering and extended class label set |
publisher |
Springer |
publishDate |
2023 |
url |
https://link.springer.com/article/10.1007/s11042-022-14021-5 https://dlib.phenikaa-uni.edu.vn/handle/PNK/8333 |
_version_ |
1764268034994733056 |
score |
8.891145 |