Abhirup Ghosh

Abhirup Ghosh is an Assistant Professor at the School of Computer Science, University of Birmingham, and a visiting researcher at the Mobile Systems Research Lab, University of Cambridge. His research focuses on distributed machine learning, particularly Federated Learning and Gossip Learning, applied to mobile health and mobility analysis. He holds a PhD from the University of Edinburgh and has held roles at Imperial College London and Intel Inc.

Education:

• PhD in Computer Science, University of Edinburgh (2019)
• M.Tech in Computer Science, Indian Institute of Technology Bombay (2011)
• Bachelor in Information Technology, Jadavpur University (2009)

Research Interests: Distributed Machine Learning, Privacy-Preserving Algorithms, Mobile Health, and Mobility Analysis. His work emphasizes collaborative learning on edge devices while addressing resource constraints and privacy concerns. Recent projects include early Alzheimer’s detection using mobility data and federated learning for health diagnostics.

Publications Trends: His work spans theoretical advancements (e.g., Gossip Learning convergence) and applied healthcare solutions (e.g., Alzheimer’s detection via outdoor mobility). Key areas include federated learning optimizations, privacy techniques, and domain generalization in activity recognition.

Awards:

• Best Publication of the Year (2022) from University of Cambridge’s Department of Computer Science

Lab & Collaborations: Collaborates with the Mobile Systems Research Lab at Cambridge on projects like MEDEA (Wellcome Trust-funded Alzheimer’s detection initiative). Leads efforts in cross-device learning and health data privacy.