Federated learning for image applications
With the development of artificial intelligence and deep learning, data privacy and security have become very important issues. For institutions or individuals to train the model together while ensuring data privacy, the concept of federated learning is proposed. In this project, we studied the aggr...
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Final Year Project (FYP) |
| Language: | English |
| Published: |
Nanyang Technological University
2021
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/150341 |
| _version_ | 1826128484740628480 |
|---|---|
| author | Cao, Shuxin |
| author2 | Wen Bihan |
| author_facet | Wen Bihan Cao, Shuxin |
| author_sort | Cao, Shuxin |
| collection | NTU |
| description | With the development of artificial intelligence and deep learning, data privacy and security have become very important issues. For institutions or individuals to train the model together while ensuring data privacy, the concept of federated learning is proposed. In this project, we studied the aggregation algorithm FedAvg of federated learning and applied it to different models and image data. A FedAvg framework was built for further research. To summarize the emergence of the problem and the optimization direction, we also proposed the hidden dangers of privacy leakage in federated learning. We practice membership inference on federated learning and proposed a new attack algorithm SVDD-MI with higher accuracy compared with the previous attack work on a single model. Besides, we also give up ideas to some of the defense models. Lastly, we found that the gradient of the federated learning model has the problem of leaking the privacy of the original image and successfully reconstructed part of the training image by inverting gradients. Moreover, we propose some of the defense methods which perform a good result. |
| first_indexed | 2024-10-01T07:25:42Z |
| format | Final Year Project (FYP) |
| id | ntu-10356/150341 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2024-10-01T07:25:42Z |
| publishDate | 2021 |
| publisher | Nanyang Technological University |
| record_format | dspace |
| spelling | ntu-10356/1503412023-07-07T18:20:38Z Federated learning for image applications Cao, Shuxin Wen Bihan School of Electrical and Electronic Engineering bihan.wen@ntu.edu.sg Engineering::Electrical and electronic engineering With the development of artificial intelligence and deep learning, data privacy and security have become very important issues. For institutions or individuals to train the model together while ensuring data privacy, the concept of federated learning is proposed. In this project, we studied the aggregation algorithm FedAvg of federated learning and applied it to different models and image data. A FedAvg framework was built for further research. To summarize the emergence of the problem and the optimization direction, we also proposed the hidden dangers of privacy leakage in federated learning. We practice membership inference on federated learning and proposed a new attack algorithm SVDD-MI with higher accuracy compared with the previous attack work on a single model. Besides, we also give up ideas to some of the defense models. Lastly, we found that the gradient of the federated learning model has the problem of leaking the privacy of the original image and successfully reconstructed part of the training image by inverting gradients. Moreover, we propose some of the defense methods which perform a good result. Bachelor of Engineering (Electrical and Electronic Engineering) 2021-06-13T12:21:40Z 2021-06-13T12:21:40Z 2021 Final Year Project (FYP) Cao, S. (2021). Federated learning for image applications. Final Year Project (FYP), Nanyang Technological University, Singapore. https://hdl.handle.net/10356/150341 https://hdl.handle.net/10356/150341 en application/pdf Nanyang Technological University |
| spellingShingle | Engineering::Electrical and electronic engineering Cao, Shuxin Federated learning for image applications |
| title | Federated learning for image applications |
| title_full | Federated learning for image applications |
| title_fullStr | Federated learning for image applications |
| title_full_unstemmed | Federated learning for image applications |
| title_short | Federated learning for image applications |
| title_sort | federated learning for image applications |
| topic | Engineering::Electrical and electronic engineering |
| url | https://hdl.handle.net/10356/150341 |
| work_keys_str_mv | AT caoshuxin federatedlearningforimageapplications |