PhD Position in Systems and Control Theory for Energy-based Learning

The University of Groningen is offering a fully-funded PhD position focused on systems and control theory for learning in neuromorphic circuits. The project aims to establish a novel system-theoretic framework that models learning as a feedback interconnection of continuous-time circuit dynamics and optimization algorithms. The successful candidate will design energy-based learning algorithms, analyze dissipative networks, and derive rigorous convergence guarantees. The position provides an opportunity to contribute to brain-inspired computing paradigms that reduce energy consumption.

• Develop a system-theoretic framework modeling learning as feedback interconnection between continuous-time circuit dynamics and optimization algorithms.
• Design novel energy-based learning algorithms for training analog circuits directly from input-output data.
• Develop fully decentralized learning rules relying on local circuit information suitable for large-scale systems.
• Establish rigorous theoretical guarantees for convergence and scalability of the proposed learning algorithms.
• Extend the mathematical theory from analog circuits to more general dissipative networks.
• Test and validate the developed methods through simulation or analytical approaches.
• Publish research findings in leading international journals and conferences and present work at scientific meetings.
• Contribute to teaching activities and supervise Bachelor's and Master's students where appropriate.