I am a first-year Ph.D. student at McGill University and Mila – Quebec AI Institute, co-advised by Prof. Doina Precup and Prof. Xiao-Wen Chang. My research focuses on continual reinforcement learning, aiming to develop AI systems that can learn and adapt in non-stationary environments. I am also a Board Member of ContinualAI, a nonprofit organization advancing research in continual learning.
Previously, I completed my M.Sc. in Electrical and Computer Engineering at McGill and Mila, under the supervision of Prof. Narges Armanfard and Prof. Samira Ebrahimi Kahou, where I worked on anomaly detection. I also hold a B.Sc. in Electrical and Computer Engineering, with a Minor in Mathematics, from the University of Alberta.
") does not match the recommended repository name for your site ("
").
", so that your site can be accessed directly at "http://
".
However, if the current repository name is intended, you can ignore this message by removing "{% include widgets/debug_repo_name.html %}
" in index.html
.
",
which does not match the baseurl
("
") configured in _config.yml
.
baseurl
in _config.yml
to "
".
Zihan Wang, Samira Ebrahimi Kahou, Narges Armanfard
Proceedings of the British Machine Vision Conference (BMVC) 2025 Oral
Zero-shot anomaly detection (ZSAD) aims to identify and localize unseen defects without requiring any labeled anomalies, but existing methods struggle to generalize under domain shifts. We propose PILOT, a framework combining a dual-branch prompt learning mechanism with label-free test-time adaptation, enabling dynamic adaptation to new distributions using only unlabeled data. PILOT achieves state-of-the-art performance on 13 industrial and medical benchmarks for both anomaly detection and localization under domain shift.
Zihan Wang, Samira Ebrahimi Kahou, Narges Armanfard
Proceedings of the British Machine Vision Conference (BMVC) 2025 Oral
Zero-shot anomaly detection (ZSAD) aims to identify and localize unseen defects without requiring any labeled anomalies, but existing methods struggle to generalize under domain shifts. We propose PILOT, a framework combining a dual-branch prompt learning mechanism with label-free test-time adaptation, enabling dynamic adaptation to new distributions using only unlabeled data. PILOT achieves state-of-the-art performance on 13 industrial and medical benchmarks for both anomaly detection and localization under domain shift.
Weijie Sun, Sunil Vasu Kalmady, Nariman Sepehrvand, Luan Manh Chu, Zihan Wang, Amir Salimi, Abram Hindle, Russell Greiner, Padma Kaul
NeurIPS 2022 Workshop on Learning from Time Series for Health 2022
We show that pre-training deep learning models on pre-pandemic health records and fine-tuning them with limited pandemic data can substantially improve ECG-based COVID-19 diagnosis and prognosis. This transfer learning approach demonstrates notable gains across three prediction tasks, highlighting its potential for rapid AI deployment in future outbreaks.
Weijie Sun, Sunil Vasu Kalmady, Nariman Sepehrvand, Luan Manh Chu, Zihan Wang, Amir Salimi, Abram Hindle, Russell Greiner, Padma Kaul
NeurIPS 2022 Workshop on Learning from Time Series for Health 2022
We show that pre-training deep learning models on pre-pandemic health records and fine-tuning them with limited pandemic data can substantially improve ECG-based COVID-19 diagnosis and prognosis. This transfer learning approach demonstrates notable gains across three prediction tasks, highlighting its potential for rapid AI deployment in future outbreaks.
Zihan Wang, Ruimin Chen, Mengxuan Liu, Guanfang Dong, Anup Basu
International Conference on Smart Multimedia 2022
We propose SPGNet, a method for 3D human pose estimation that integrates multi-dimensional re-projection into supervised learning. Our approach enforces kinematic constraints and jointly optimizes both 2D and 3D pose consistency, leading to improved accuracy. Experiments on the Human3.6M dataset show that SPGNet outperforms many state-of-the-art methods.
Zihan Wang, Ruimin Chen, Mengxuan Liu, Guanfang Dong, Anup Basu
International Conference on Smart Multimedia 2022
We propose SPGNet, a method for 3D human pose estimation that integrates multi-dimensional re-projection into supervised learning. Our approach enforces kinematic constraints and jointly optimizes both 2D and 3D pose consistency, leading to improved accuracy. Experiments on the Human3.6M dataset show that SPGNet outperforms many state-of-the-art methods.