BKK19-402 - Inferencing at the edge and Fragmentation Challenges

Session Abstract

As deep learning (DL) expands is application into ever more areas, DL at the edge has become an area of rapid innovation and has also become highly fragmented. This creates a challenge in the ecosystem for framework providers that want to take advantage of specialized hardware, and an equal challenge for SoC providers, or makers of DL accelerators that need to support various frameworks, customer innovations, device constraints, etc. This talk will explore what constitutes DL at the edge, it will highlight the recent trends in this area from runtimes and compilers, to model formats, and explore the challenges, and scalability needs of collaborative solutions.

Session Speakers

Mark Charlebois

Director, Engineering Qualcomm Technologies Inc (Qualcomm Technologies Inc)

Mark is Director Engineering in Qualcomm Technologies Inc (QTI) in the Machine Learning Group. Currently he is focused on Neural Processing Runtime for Qualcomm SoCs, AI Benchmarking, and also serves as an open source Trusted Advisor for the MLG group. He has represented QTI on the Linux Foundation board, served on the Dronecode board and Core Infrastructure Initiative steering committee, and as the TSC lead for Dronecode. Mark also contributed code to the PX4 Open Source Flight Stack (http://github.com/PX4/Firmware), and to the LLVMLinux project with associated patches for the Linux kernel. He also helps support the Dragonboard developer platforms and has been working in embedded software for over 25 years. Mark has a BASc in Systems Design Engineering from the University of Waterloo, and a MASc in Engineering Science from Simon Frazer University.

comments powered by Disqus

Other Posts

Sign up. Receive Updates. Stay informed.

Sign up to our mailing list to receive updates on the latest Linaro Connect news!