positions

Task: Exploring uncertainty quantification of on-device learning for health (WP4)

Host institution: Cambridge

Country: United Kingdom

Supervisor: Prof. C. Mascolo [Cambridge]

Co-supervisors: Prof. Q. Wang [TU Delft]; Dr. M. Petković [Philips]

Objectives: 1) To design the next-generation uncertainty-aware machine learning models optimised for on device computation in the context of streaming and longitudinal data analysis; 2) To integrate the developed approaches with continual learning to allow for the models to
continually learn and avoid cata-strophic forgetting; 3) To evaluate the proposed solutions using existing wearable and health datasets and implementation and testing on real microcontroller boards to refine the systems performance and efficiency.

Expected Results: 1) Novel efficient uncertainty-aware models for microcontroller units (MCUs) targeting longitudinal wearable data. 2) Continual learning uncertainty-aware MCU model for longitudinal wearable data.

PhD enrolment: Doctoral School of Cambridge

Planned secondments: 

• TU Delft (3 months, M16-M18): Uncertainty-aware machine learning models optimised for on-device computing, with Prof. Q. Wang (KPI: joint paper)

• Philips (3 months, M28-M30): Proof-of-concept evaluation using real-life wearable and health datasets and real microcontroller boards, with Dr. M. Petković (KPI: joint paper)

Candidate profile: computer science, electrical engineering, biomedical engineering (in order of preference)

Desirable skills/interests: on-device machine learning, wearable systems, signal processing, statistical filtering, machine learning applied to health data (the applicant should be proficient in at least one or two of the skills)

Application Deadline: February 14, 2025, AoE

Descriptions of all the 18 DC Positions

Submit Your Application HERE