Recruiting
We're actively recruiting motivated undergraduate, Master's, and Ph.D. students to join HUB Lab. If you're interested in AI, deep learning, wearable sensing, AI for healthcare, or time series analysis, please send us your CV and a short note about your research interests. HUB Lab 正在招收有熱忱的大學部、碩士班與博士班同學。如果你對人工智慧(AI)深度學習穿戴式感測健康照護分析,或時間序列分析等方向有興趣,歡迎將履歷與一段對研究方向的想法寄給我們。
Contact Us
If you're applying for a Master's degree, please apply to the Institute of Computer Science and Engineering (資訊科學與工程研究所-甲組) or the Institute of Data Science and Engineering (數據科學與工程研究所) at NYCU. Once you receive an admission offer, please feel free to get in touch. I'd love to hear about your research interests and explore how we might work together. 如果你想申請碩士班,可以報考國立陽明交通大學的資訊科學與工程研究所-甲組數據科學與工程研究所。收到錄取通知後,歡迎隨時來信,跟我聊聊你的研究興趣,以及我們可以一起進行的研究。
Human Understanding & Behavior Lab

Po-Kai Yang

Incoming Assistant Professor (Tenure-Track), Department of Computer Science
National Yang Ming Chiao Tung University (NYCU)
國立陽明交通大學 資訊工程學系
專任助理教授(預計 2026 年 8 月到任)

Po-Kai Yang

Our Vision

We build multimodal AI systems that decode human behavior from everyday sensor data, enabling earlier diagnosis, personalized rehabilitation, and adaptive human-machine interaction across healthcare and immersive computing. 本實驗室發展能整合多種感測訊號的分析方法,從日常生活的感測資料解讀人類的行為與動作。研究目標是支援更早期的疾病診斷、個人化的復健介入,以及更貼近使用者的人機互動,應用範圍涵蓋健康照護以及沉浸式運算。

Research Directions

Our near-term research builds multimodal AI for behavioural health: the clinical foundation for broader human understanding and interaction. 實驗室目前的研究結合不同類型的感測訊號,用於分析人類行為與健康狀態;這些臨床導向的工作,也是我們在人類理解與人機互動研究上的基礎。

Multimodal Human Behavior Understanding

Inferring both explicit intentions and implicit states by fusing motion, cardiac-autonomic, speech, and gaze signals from wearable and ambient sensors. 整合穿戴式與環境中各類感測器所收集的訊號(包括身體動作、心律與自律神經反應、語音、眼動等),同時理解使用者外顯的意圖以及不易察覺的內在狀態。

IMU & Depth ECG & PPG Speech & Prosody Eye Gaze

Clinical Digital Biomarkers

Developing and validating cross-modal digital biomarkers for neurological and mental health conditions, in collaboration with hospital clinicians. 與醫院的臨床團隊合作,發展並驗證能整合多種訊號來源的數位生物標記,用於神經系統疾病以及心理健康狀態的客觀評估。

Parkinson's Disease Stroke Depression Clinical Validation

Multimodal Foundation Models

Building cross-modal behaviour foundation models with layered pre-training, interpretability mechanisms, and open-source benchmarking for clinical deployment. 發展能整合多種感測訊號的行為分析模型基礎,研究階段性的學習機制以及模型可解釋性,並透過開放原始碼資料集與基準測試,支援後續的臨床應用。

Cross-Modal Learning Foundation Model Open-Source Interpretability
Selected Works

Highlighted papers and recent publications. See Google Scholar for the full list. 代表性論文與近期發表。完整著作清單請見 Google Scholar

Courses

Bridging theory and hands-on practice in AI, sensor systems, and human behavior understanding. 結合理論與實作,內容涵蓋資料分析方法、感測系統,以及人類行為的理解。

Course offerings

To be announced

Courses for the upcoming academic year are being finalized. Details will be posted here once confirmed by the department. 下一學年度的開課內容仍在規劃中。系所確認後將於此處公告。

Team

Building a collaborative, multidisciplinary research team.

Po-Kai Yang

Po-Kai Yang

Principal Investigator

PhD Students

Recruiting now

Master's Students

Recruiting now

Undergraduate

Recruiting now

Collaborators
World map
Taiwan
HUB Lab (NYCU)
Academic & Industry Partners
News

Recent announcements from the lab.

Sep 2026

Organizing special session at EUSIPCO 2026 Upcoming Conference

Co-organizing "Advanced Time-Series Modeling for Wearable-Based ADL Monitoring in Healthcare" at EUSIPCO 2026 in Bruges, Belgium.

Aug 2026

Joining NYCU CS as Tenure-Track Assistant Professor Upcoming Career

Po-Kai Yang joins the Department of Computer Science at National Yang Ming Chiao Tung University (NYCU), Taiwan (R.O.C.) as a Tenure-Track Assistant Professor.

Aug 2026

HUB Lab established at NYCU Upcoming Lab

The Human Understanding & Behavior Lab officially launches in the Department of Computer Science.

Sep 2025

PhD thesis defended Milestone

Successfully defended PhD thesis on "Automated Detection and Assessment of Freezing of Gait in Parkinson's Disease using Wearable Sensors and Machine Learning" at KU Leuven.

Jun 2025

Best Poster Award at ICFOG 2025 Award

Our work on deep learning-based assessment of freezing of gait severity wins the Best Poster Award at the International Conference on Freezing of Gait in Atlanta.

2025

Most Cited Paper in JNER 2024 Paper

Our FOG assessment paper in the Journal of NeuroEngineering and Rehabilitation is recognised as the most cited paper of 2024.

Join HUB Lab

We have open positions for motivated undergraduate, Master's, and PhD students interested in AI, deep learning, multimodal sensing, clinical neuroscience, and human-computer interaction. We value curiosity, initiative, and collaborative spirit. 實驗室正在招收博士班、碩士班與大學部專題生。我們研究人工智慧(AI)與深度學習,應用於穿戴式感測訊號分析、臨床健康照護、臨床神經科學,以及人機互動。歡迎對任何方向有興趣的同學與我們聯繫;我們重視好奇心、主動學習,以及跨領域合作的開放態度。

Office: EC Building, Room TBD · NYCU, Hsinchu, Taiwan 辦公室:國立陽明交通大學 工程三館(待確認)· 新竹

Get in Touch

Send your CV and a brief research statement to po-kai.yang@kuleuven.be 請將履歷及簡短的研究興趣說明寄至 po-kai.yang@kuleuven.be

Visit Us

By appointment during office hours · email to schedule 辦公時間歡迎來訪,請先以電子郵件預約

English-Friendly Lab

We encourage English communication. Students are supported to attend international conferences and visit partner labs abroad 實驗室以中英雙語溝通,並支持同學參加國際會議與海外合作實驗室短期訪問