2019 International Conference On Computer Aided Design

The Premier Conference Devoted to Technical Innovations in Electronic Design Automation

November 4-7, 2019The Westin Westminster Westminster, CO

v class="event-details"> MP Associates, Inc.

WEDNESDAY November 07, 3:45pm - 5:15pm | Capri

A Hardware Outlook on Machine Learning Applications
Da-Cheng Juan - Google, Inc.
This session brings together perspectives from industrial and academic viewpoints on the importance of
hardware architecture choice on machine learning applications. The first talk addresses the importance
of hardware‐aware optimized software kernels in achieving energy‐ and resource‐efficient deployment of
neural networks (NNs) on IoT edge devices. By augmenting classic NN architecture search techniques with hardware parameters, the second talk addresses the topic of achieving Pareto optimal hardware‐aware model configurations for both high‐end and mobile platforms. Concluding the session, the third talk discusses hardware‐aware power and runtime modeling and optimization techniques that enable NN-hardware platform co‐design exploration.

10D.1Enabling Deep Learning at the IoT Edge
 Speaker: Naveen Suda - Arm, Ltd.
 Authors: Liangzhen Lai - Arm, Ltd.
Naveen Suda - Arm, Ltd.
10D.2Searching Toward Pareto‐Optimal Device‐Aware Neural Architectures
 Speaker: Da-Cheng Juan - Google, Inc.
 Author: Da-Cheng Juan - Google, Inc.
10D.3Hardware‐Aware Machine Learning: Modeling and Optimization
 Speaker: Diana Marculescu - Carnegie Mellon Univ.
 Authors: Diana Marculescu - Carnegie Mellon Univ.
Dimitrios Stamoulis - Carnegie Mellon Univ.
Ermao Cai - Carnegie Mellon Univ.