BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Energy Efficient Edge AI using Ensembles and Hyperdimensional Comp
 uting - Flavio Ponzina\, UCSD
DTSTART:20240916T123000Z
DTEND:20240916T133000Z
UID:TALK220858@talks.cam.ac.uk
CONTACT:Nic Lane
DESCRIPTION:The growing adoption of artificial intelligence (AI) across a 
 wide range of applications is accompanied by a corresponding increase in m
 odel complexity\, which introduces new challenges for deploying AI on edge
  devices with constrained memory\, limited computational capacity\, and st
 rict energy requirements. Research efforts in this area are focused on bot
 h algorithmic and hardware optimization\, often balancing accuracy against
  energy efficiency. At the algorithmic level\, Hyperdimensional Computing 
 (HDC) has emerged as a promising approach to either replace or complement 
 conventional models such as deep neural networks\, while processing-in-mem
 ory is increasingly recognized as a well-established strategy for hardware
  acceleration. In my work\, I propose efficient hardware-software co-desig
 n methodologies and model optimizations that enable highly accurate edge A
 I while simultaneously enhancing computational efficiency. I demonstrate t
 he effectiveness of hardware-aware ensembling methods for edge AI\, highli
 ghting their exceptional efficiency when integrating HDC learners. Further
 more\, I exploit the significant parallelism inherent in HDC classifiers\,
  co-designing them with analog memory arrays for processing-in-memory (PIM
 ) acceleration.\n\nBio\n\nFlavio Ponzina earned his M.Sc. degree in Comput
 er Engineering from Politecnico di Torino\, Italy\, in 2018\, and his Ph.D
 . degree in Electronic Engineering from EPFL\, Switzerland\, in 2023. He i
 s currently a postdoctoral scholar at the University of California\, San D
 iego (UCSD)\, La Jolla\, CA\, USA. His doctoral research focused on hardwa
 re-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\, he is expanding this work by expl
 oring the co-optimization of hardware and software\, with an emphasis on H
 yperdimensional Computing (HDC)\, a brain-inspired computational paradigm\
 , and emerging memory technologies.\n\n
LOCATION:Computer Lab\, FW26
END:VEVENT
END:VCALENDAR