| 課程編碼 Course Code | 中文課程名稱 Course Name (Chinese) | 英文課程名稱 Course Name (English) | 總學分數 Credits | 總時數 Hours |
|---|---|---|---|---|
| 3604178 | 智慧醫療資料自主與聯邦式學習 | Smart Healthcare Data Autonomy and Federated Learning | 3.0 | 3 |
| 中文概述 Chinese Description | 本課程以智慧醫療應用為核心場域，聚焦聯邦式學習（Federated Learning）之演算法架構與實務工程實作，探討在資料不集中前提下之分散式模型訓練、參數聚合策略、通訊效率優化與系統效能評估。課程強調跨場域醫療資料隱私保護機制，包括差分隱私、安全聚合與模型更新穩定性設計，並透過實作專案訓練學生建構可部署之跨醫療機構 AI 系統。教學內容以演算法推導、實驗設計與工程實作為主軸，不涉及資料結構理論，培養學生具備分散式醫療 AI 系統設計與效能優化能力。 | |||
| 英文概述 English Description | This course focuses on algorithmic design and practical implementation of Federated Learning in smart healthcare applications. It addresses distributed model training, aggregation strategies, communication efficiency, and system performance evaluation under non-centralized data settings. Emphasis is placed on privacy-preserving mechanisms such as differential privacy, secure aggregation, and robust model updating in cross-institutional medical AI systems. Students will engage in hands-on engineering projects to develop deployable federated healthcare AI frameworks. The course centers on algorithms, experimental validation, and system-level implementation rather than data structure theory, aiming to cultivate the ability to design efficient and privacy-aware distributed medical AI systems. | |||
備註：