Session

· Contributors · Organizations · Search Program · My Agenda · Happening Now

Research Manuscript: Pushing the Boundaries of Practical AI: Efficiency and Robustness

Session Chairs

Haoyu Yang

NVIDIA

Yingyan Lin

Rice University

Event TypeResearch Manuscript

Keywords

Topics

TimeWednesday, July 13th1:30pm - 3:00pm PDT

Location3000, Level 3

DescriptionWhile machine learning has recently achieved various performance breakthroughs, its learning efficiency and robustness are critical to more extensive applications. This session presents three efficient learning techniques for quantum neural networks, edge computing, and logic representation, respectively. The last paper introduces a method for better fault tolerance.

Presentations

1:30pm - 1:52pm PDT	QOC: Quantum On-Chip Training with Parameter Shift and Gradient Pruning Authors Hanrui Wang Zirui Li Jiaqi Gu Yongshan Ding David Z. Pan Song Han
1:52pm - 2:15pm PDT	Memory-Efficient Training of Binarized Neural Networks on the Edge Authors Mikail Yayla Jian-Jia Chen
2:15pm - 2:37pm PDT	DeepGate: Learning Neural Representations of Logic Gates* Authors Min Li Sadaf Khan Zhengyuan Shi Naixing Wang Yu Huang Qiang Xu
2:37pm - 3:00pm PDT	Bipolar Vector Classifier for Fault-tolerant Deep Neural Networks Authors Suyong Lee Insu Choi Joon-Sung Yang