positions

Task: Split Learning across underwater and surface devices (WP4)

Host institution: UNITN

Country: Italy

Supervisor: Prof. P. Casari [UNITN]

Co-supervisors: Prof. D. Ganesan [UMass]; Dr. E. Rocco [A5]

Objectives: 1) To investigate split-computing neural network architectures that strike an optimal balance amongst the computing complexity, data volume, and energy consumption; 2) To design flexible optimisation and early exiting techniques that adapt to different communication technologies (e.g., acoustic in different bandwidths, optical); 3) To conduct proof-of-concept experiments using underwater devices.

Expected Results: 1) A split-computing architecture that caters for the energy constrains of underwater devices and minimises data transfers; 2) Flexible optimisation and early exiting techniques; 3) Proof-of-concept experiments showing real-time mission adaptation based on the outcome of split computing.

PhD enrolment: Doctoral School of UNITN

Planned secondments: 

• UMass (3 months, M16-M18): Split computing architecture and early-exit techniques under energy constraints, with Prof. D. Ganesan (KPI: joint paper)

• A5 (3 months, M28-M30): Proof-of-concept split-computing solutions for underwater image classification and anomaly detection tasks, with Dr. E. Rocco (KPI: joint paper)

Candidate profile: computer science, telecommunication engineering, data science, applied mathematics, electrical engineering, (in order of preference)

Desirable skills/interests: signal processing, statistical filtering, machine learning and deep learning, split computing architectures, applied optimization (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