🔬 A Glance at Emilio Paolini and His Fellowship
Meet the Researcher
Emilio Paolini is an Assistant Professor and researcher at Scuola Superiore Sant’Anna in Italy, where he focuses on developing efficient, next-generation machine learning systems for wireless networks. With a Ph.D. in Emerging Digital Technologies—centered on AI techniques for constrained environments such as neuromorphic photonic accelerators—and degrees in Artificial Intelligence, Data Engineering, and Computer Engineering, Emilio brings a rich academic foundation to the evolving field of digital communications. His research spans network intelligence, neuromorphic computing, and real-time AI optimization, all aimed at enhancing the performance and sustainability of wireless systems.
The Fellowship and Transatlantic Collaboration
Thanks to the support of the NGI Enrichers fellowship, Emilio launched the project Adaptive Edge-AI Deployment in NextG Wireless Networks, a forward-looking initiative that tackles one of the central challenges of modern wireless infrastructures: how to deploy complex deep learning models in energy-efficient, low-latency ways under ever-changing network conditions and heterogeneous hardware setups.
The project combines advanced techniques—such as compression, pruning, quantization, and adaptive neural architecture search—to dynamically adapt AI models in real time. It integrates photonic acceleration to speed up computation and federated learning to enable decentralized model training with reduced communication overhead and improved privacy. These innovations are validated in a joint testbed environment developed collaboratively by Scuola Superiore Sant’Anna and Saint Louis University in Missouri.
Emilio chose Saint Louis University as his host institution due to its strong expertise in computer science, distributed edge-cloud systems, and network management. Under the mentorship of Dr. Flavio Esposito, whose research aligns closely with the project’s vision, the partnership created a dynamic research environment that bridges academic rigor with practical, industry-relevant experimentation.
This transatlantic collaboration, made possible through NGI Enrichers, exemplifies how international research exchange can catalyze innovation in real-world technologies, pushing forward the boundaries of what’s possible in adaptive, AI-driven wireless systems.
đź“„ Scientific Highlight: Flecto
One of the key outputs of Emilio’s fellowship is his scientific publication, Flecto: Cross-Layer Adaptive Congestion Control with Reinforcement Learning. Developed within the NGI Enrichers framework, Flecto addresses the pressing need for smarter congestion control mechanisms in next-generation networks.
By incorporating reinforcement learning into the QUIC protocol and analyzing data across multiple network layers—from physical signal quality to transport-level indicators—Flecto dynamically adjusts transmission rates to optimize performance in real time. The results speak for themselves: an average throughput of 4539.5 KB/s, about 6% higher than Cubic, and a stable round-trip time of 21.8 ms, outperforming traditional methods such as New Reno.
More than a technical achievement, Flecto reflects the broader mission of NGI Enrichers: to empower researchers to create scalable, adaptive, and efficient digital solutions that serve both academic and industry needs.