Back to all members...
Hao Fei
Postdoc, started 2026
Hao is a senior postdoctoral researcher at OATML working with Yarin Gal, jointly affiliated at Department of Computer Science and Big Data Institute, University of Oxford. Previously, he was a senior research fellow at the National University of Singapore. His research focuses on multimodal foundational models and generative models, with an emphasis on exploring the relationship between the physical world and the mental world. Recently, he has been working on AI for natural science, AI scientific discovery. He has been continuously reflecting on 3 fundamental questions: 1) How can AI systems achieve seamless generalization across heterogeneous modalities? 2) How can AI develop human-like sensitivity to subtle cognitive and affective state changes? 3) How can AI autonomously identify and expand the boundaries of scientific knowledge under fundamentally underspecified conditions?
Publications while at OATML • News items mentioning Hao Fei • Reproducibility and Code • Blog Posts
Publications while at OATML:
PRISM-CTG: A Foundation Model for Cardiotocography Analysis with Multi-View SSL
Supervised deep learning models for automated CTG analysis are typically constrained by narrowly curated labelled datasets and limited patient cohorts, leaving substantial volumes of physiologically informative clinical recordings untapped. To address this limitation, we propose Physiology-aware Representation Learning via Integrated Self-supervision and Metadata for CTG (PRISM-CTG), a clinically grounded self-supervised foundation model (FM) for CTG that leverages large-scale unlabelled recordings to learn transferable domain-level representations. PRISM-CTG is pretrained using a multi-view self-supervised framework that jointly optimises 3 complementary pretext objectives: random-projected guided masked signal reconstruction, clinical variable prediction, and feature classification. Each objective is associated with a dedicated task-specific token, enabling specialised representation learning, while controlled cross-attention facilitates information exchange across clinical conte... [full abstract]
Sheng Wong, Ravi Shankar, Beth Albert, Hao Fei, Lin Li, Imane Ben M'Barek, Manu Vatish, Gabriel Jones
arxiv
[paper]