On January 16, 2026, Vishnu Moothedath successfully defended his PhD thesis entitled “Towards Efficient Distributed Intelligence: Cost-Aware Sensing and Offloading for Inference at the Edge.” The defence marked an important contribution to research on distributed intelligence in the context of AI-driven systems and future wireless networks.
The thesis addresses fundamental challenges arising from the increasing deployment of intelligent systems operating under strict latency and energy constraints. Rather than focusing solely on inference accuracy, the work investigates how decisions about when to sense the environment and when to offload computation can be optimised using cost-aware and semantics-driven approaches. By keeping the formulation platform-independent and relying on pre-trained models, the results are applicable across a wide range of distributed and edge-computing systems.
The opponent at the defence was Professor Ayalvadi Ganesh (University of Bristol, UK). The examination committee consisted of Dr. Beatriz Grafulla (Ericsson), Associate Professor Praveen Kumar Donta (Stockholm University), and Associate Professor Salman Toor (Uppsala University), with Professor Rolf Stadler (KTH) serving as stand-in member. The defence was chaired by Mikael Skoglund (KTH).
Vishnu Moothedath was supervised by Professor James Gross and Professor György Dan. The thesis advances the state of the art in efficient distributed inference and provides a foundation for future research on intelligent edge systems.
