Adversarial attacks on deep learning
Deep learning models, especially convolutional neural networks (CNNs), have made significant progress in the field of image recognition and classification. However, adversarial attacks have emerged as a significant vulnerability, posing threats to the robustness of these models. One notable example...
| 1. autor: | |
|---|---|
| Kolejni autorzy: | |
| Format: | Final Year Project (FYP) |
| Język: | English |
| Wydane: |
Nanyang Technological University
2023
|
| Hasła przedmiotowe: | |
| Dostęp online: | https://hdl.handle.net/10356/166036 |
| _version_ | 1826129368906203136 |
|---|---|
| author | Yee, An Qi |
| author2 | Lam Siew Kei |
| author_facet | Lam Siew Kei Yee, An Qi |
| author_sort | Yee, An Qi |
| collection | NTU |
| description | Deep learning models, especially convolutional neural networks (CNNs), have made significant progress in the field of image recognition and classification. However, adversarial attacks have emerged as a significant vulnerability, posing threats to the robustness of these models. One notable example is the one-pixel attack, which leads to incorrect predictions just by changing a single pixel, which could lead to potentially serious consequences. This project aims to investigate the efficiency and effectiveness of different search strategies in conducting the one- pixel attacks on black box networks.
Certain adversarial attacks are explored before narrowing down to one pixel attack. This study will further explore the performance of three search algorithms - Genetic Algorithm (GA), Simulated Annealing (SA) and Differential Evolution (DE) - in terms of the computational power used, success rates and convergence speed. The aim of this study is to research on the effects of these algorithms on one pixel attack, hopefully achieving the goal to identify elements that improve the efficiency and efficacy of the one-pixel attack. |
| first_indexed | 2024-10-01T07:39:26Z |
| format | Final Year Project (FYP) |
| id | ntu-10356/166036 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2024-10-01T07:39:26Z |
| publishDate | 2023 |
| publisher | Nanyang Technological University |
| record_format | dspace |
| spelling | ntu-10356/1660362023-04-21T15:39:37Z Adversarial attacks on deep learning Yee, An Qi Lam Siew Kei School of Computer Science and Engineering Li Yi ASSKLam@ntu.edu.sg, yi_li@ntu.edu.sg Engineering::Computer science and engineering Science::Mathematics Deep learning models, especially convolutional neural networks (CNNs), have made significant progress in the field of image recognition and classification. However, adversarial attacks have emerged as a significant vulnerability, posing threats to the robustness of these models. One notable example is the one-pixel attack, which leads to incorrect predictions just by changing a single pixel, which could lead to potentially serious consequences. This project aims to investigate the efficiency and effectiveness of different search strategies in conducting the one- pixel attacks on black box networks. Certain adversarial attacks are explored before narrowing down to one pixel attack. This study will further explore the performance of three search algorithms - Genetic Algorithm (GA), Simulated Annealing (SA) and Differential Evolution (DE) - in terms of the computational power used, success rates and convergence speed. The aim of this study is to research on the effects of these algorithms on one pixel attack, hopefully achieving the goal to identify elements that improve the efficiency and efficacy of the one-pixel attack. Bachelor of Science in Mathematical and Computer Sciences 2023-04-19T08:27:31Z 2023-04-19T08:27:31Z 2023 Final Year Project (FYP) Yee, A. Q. (2023). Adversarial attacks on deep learning. Final Year Project (FYP), Nanyang Technological University, Singapore. https://hdl.handle.net/10356/166036 https://hdl.handle.net/10356/166036 en SCSE22-0150 application/pdf Nanyang Technological University |
| spellingShingle | Engineering::Computer science and engineering Science::Mathematics Yee, An Qi Adversarial attacks on deep learning |
| title | Adversarial attacks on deep learning |
| title_full | Adversarial attacks on deep learning |
| title_fullStr | Adversarial attacks on deep learning |
| title_full_unstemmed | Adversarial attacks on deep learning |
| title_short | Adversarial attacks on deep learning |
| title_sort | adversarial attacks on deep learning |
| topic | Engineering::Computer science and engineering Science::Mathematics |
| url | https://hdl.handle.net/10356/166036 |
| work_keys_str_mv | AT yeeanqi adversarialattacksondeeplearning |