Close

Session

Research Manuscript: Smarter Memory Solutions for Faster ML
Event TypeResearch Manuscript
Virtual Programs
Hosted in Virtual Platform
Keywords
AI/ML System Design
Topics
Design
Time
Location
DescriptionThe well-known memory-wall issue is the major bottleneck of data-intensive machine learning based computing. This session presents two different approaches to tackle this issue through either emerging in-memory computing architecture based on SRAM/RRAM or optimizing dynamic refresh of DRAM.