H11-M02_code.zip
From the dataset abstract
GradientHide 透過新增公共資料更新步驟並使用 CLIP 進行標籤對齊來防止聯邦學習中的梯度反轉，從而有效地保護隱私，同時保持跨基準資料集的模型準確性。
Source: H11-M02_隱私保護之聯邦式學習模型
Additional Information
| Field | Value |
|---|---|
| Data last updated | October 16, 2025 |
| Metadata last updated | October 16, 2025 |
| Created | October 16, 2025 |
| Format | application/zip |
| License | Other (Non-Commercial) |
| created | 1 month ago |
| format | ZIP |
| id | 069d28a7-68ff-49de-9210-7dfe587ee8d4 |
| last modified | 1 month ago |
| mimetype | application/zip |
| on same domain | True |
| package id | ab7ef4cd-36b2-42c6-b73c-9590c2b37ce0 |
| position | 1 |
| proxy url | https://scidm.nchc.org.tw/en/dataset/ab7ef4cd-36b2-42c6-b73c-9590c2b37ce0/resource/069d28a7-68ff-49de-9210-7dfe587ee8d4/nchcproxy/H11-M02_code.zip |
| revision id | b514f820-0285-4216-907f-c396bfe3f7bb |
| size | 130.8 MiB |
| state | active |
| url type | upload |
推薦資料集：