About me

I am Flavio Ponzina, currently a postdoctoral scholar in the Computer Science and Engineering department at University of California San Diego (UCSD) working at the Systems Energy Efficiency Lab with Prof. Tajana Rosing. Prior to joining UCSD, I was a PhD student in Electrical and Computer Engineering at the École Polytechnique Fédérale de Lausanne (EPFL), Switzerland, working at the Embedded Systems Laboratory under the supervision of Prof. David Atienza until September 2023. I received my Master of Science in Computer Engineering with high honors from Politecnico di Torino, Italy, in 2018.
My doctoral research focused on hardware-software co-design for energy-efficient edge AI, particularly through the integration of neural network-based ensembling methods with processing-in-memory (PIM) acceleration. At UCSD, I am expanding this work by exploring the co-optimization of hardware and software, with an emphasis on Hyperdimensional Computing (HDC), a brain-inspired computational paradigm, and emerging memory technologies such as RRAM, PCM, and MRAM.

Research Interests

I am interested in energy-efficient design and HW-SW co-optimization for both ultra-low-power IoT systems and large-scale applications. More specific research directions include:

Ensemble learning for accurate and robust AI

HW-SW co-design for Edge AI optimizations

Distributed unsupervised learning under constrained resources

Lifelong learning in connected IoT systems

Energy-efficient optimizations of large-scale vector databases

Teaching Experience

CSE 147. Introduction to Embedded Systems University of California San Diego, Winter 2025

CSE 237a. Introduction to Embedded Computing University of California San Diego, Winter 2025

EE-310. Microprogrammed Embedded Systems École Polytechnique Fédérale de Lausanne (EPFL), 2020-2023

Selected Publications

Le Zhang, Onat Gungor, Flavio Ponzina, and Tajana Rosing., “E-QUARTIC: Energy Efficient Edge Ensemble of Convolutional Neural Networks for Resource-Optimized Learning”, ASP-DAC, 2025

Kumar Ashwani, Yucheng Zhou, Sai Praneeth Potladurthy, Jeoghoon Kim, Weihong Xu, Flavio Ponzina, Seounghyun Kim, Ertugrul Cubukcu, Tajana Rosing, Gert Cauwenbergh, and Duygu Kuzum. “Filament-free Bulk RRAM with High Endurance and Long Retention for Neuromorphic Few-Shot Learning On-Chip”, IEDM, 2024

Flavio Ponzina, Rishikanth Chandrasekaran, Anya Wang, Seiji Minowada, Siddharth Sharma, and Tajana Rosing. “Multi-Model Inference Composition of Hyperdimensional Computing Ensembles”, ICCD, 2024

Keming Fan, Ashkan Moradifirouzabadi, Xiangjin Wu, Zheyu Li, Flavio Ponzina, Anton Persson, Vikram Adve, Eric Pop, Tajana Rosing, and Mingu Kang. “SpecPCM: A Low-power PCM-based In-Memory Computing Accelerator for Full-stack Mass Spectrometry Analysis” IEEE JXCDC, 2024.

Flavio Ponzina, Mialyssa Gomez, Congge Xu, and Tajana Rosing. “GlucoseHD Predicting Glucose Levels using Hyperdimensional Computing”, IEEE Design and Test, 2024.

Marco Rios, Flavio Ponzina, Alexandre Levisse, Giovanni Ansaloni, and David Atienza. "Bit-line computing for CNN accelerators co-design in edge AI inference" IEEE Transactions on Emerging Topics in Computing 11, no. 2, 2023.

Marco Rios, Flavio Ponzina, Giovanni Ansaloni, Alexandre Levisse, and David Atienza. "Running efficiently cnns on the edge thanks to hybrid sram-rram in-memory computing", DATE, 2021

Flavio Ponzina, Miguel Peon-Quiros, Andreas Burg, and David Atienza. "E2cnns: Ensembles of convolutional neural networks to improve robustness against memory errors in edge-computing devices" IEEE Transactions on Computers 70, no. 8, 2021.