Close

Session

Research Manuscript
:
Models and Hardware for Machine Learning and Beyond
DescriptionIn this session, we will present eight works on hardware and algorithm design for emerging computing models. These span a wide range of areas, including AI, machine learning, neuromorphic computing, bioinformatics, approximate computing, and biochips. The selected works feature novel algorithms and hardware architectures for both general AI and machine learning applications, as well as specialized AI models such as vector symbolic AI. Together, they showcase a diverse array of computing techniques, highlighting the challenges and state-of-the-art solutions within these evolving paradigms.
Event Type
Research Manuscript
TimeMonday, June 233:30pm - 5:30pm PDT
Location3003, Level 3
Topics
Design
Tracks
DES3: Emerging Models of ComputatioN
Presentations
3:30pm - 3:45pm PDTPipirima: Predicting Patterns in Sparsity to Accelerate Matrix Algebra
DES3: Emerging Models of ComputatioN
3:45pm - 4:00pm PDTFactorHD: A Hyperdimensional Computing Model for Multi-Object Multi-Class Representation and Factorization
DES3: Emerging Models of ComputatioN
4:00pm - 4:15pm PDTHolistic Design towards Resource-Stringent Binary Vector Symbolic Architecture
DES3: Emerging Models of ComputatioN
4:15pm - 4:30pm PDTSDISC: A Spike-Driven Human-Machine Interface with In-Situ Computing for Real-Time Low-Power Interaction
DES3: Emerging Models of ComputatioN
4:30pm - 4:45pm PDTANGraph: A GNN-Based Performance Prediction Framework for Asynchronous Neuromorphic Hardware
DES3: Emerging Models of ComputatioN
4:45pm - 5:00pm PDTPairGraph: An Efficient Search-space-aware Accelerator for High-performance Concurrent Pairwise Queries
DES3: Emerging Models of ComputatioN
5:00pm - 5:15pm PDTPreDAC: An Efficient Framework of Pre-Refining Enhanced Design Space Exploration for Approximate Computing
DES3: Emerging Models of ComputatioN
5:15pm - 5:30pm PDTAutoRE: Bayesian-Optimization-based Automatic Reliability Enhancement Tool for Flow-based Microfluidic Biochips
DES3: Emerging Models of ComputatioN