Hardware and Systems Engineering

Color:

#1d3459

Closing Keynote Address: Deep learning at the edge: 100x Inference Improvement on edge devices

In this presentation we discuss how leveraging sparsity, knowledge distillation and quantization, not only enable deep learning models to run on edge devices, but in many instances outperform standard models running on datacenter class systems. We discuss our recent innovations in sparsity research that enables the creation of dual-sparse networks that can outperform standard networks by over 100X. Finally, we illustrate how sparsity can be combined with other model compression techniques to unlock deep learning inference at the edge.

On Device ML

Edge Trade Offs

Innovation at the Edge

Hardware and Systems Engineering

Software Engineering

Data Science

Strategy

Author:

Lawrence Spracklen

Director of Machine Learning Architecture

Numenta

Read more about Closing Keynote Address: Deep learning at the edge: 100x Inference Improvement on edge devices

PANEL DISCUSSION: Co-design and co-optimization: maximizing energy per joule

Achieving full optimization requires co-design and co-optimization of AI algorithms and the hardware platforms where they will be executed. This panel will explore the benefits and potential of such co-design approaches.

Innovation at the Edge

Chip and Systems Design

Edge Trade Offs

On Device ML

Hardware and Systems Engineering

Software Engineering

Strategy

Author:

Subutai Ahmad

CEO

Numenta

Subutai is passionate about neuroscience, deep learning, and building intelligent systems. An accomplished technologist, he has been instrumental in driving Numenta’s research, technology and business since 2005. He previously served as VP Engineering at YesVideo where he helped grow the company from a three-person start-up to a leader in automated digital media authoring. In 1997, Subutai co-founded ePlanet Interactive which developed the IntelPlay Me2Cam, the first computer vision product developed for consumers. Subutai holds a B.S. in Computer Science from Cornell University, and a Ph.D in Computer Science from the University of Illinois at Urbana-Champaign.

Author:

Anshumali Shrivastava

CEO and Founder

ThirdAI

Author:

Godwin Maben

Strategy & System Architect and Fellow

Synopsys

Read more about PANEL DISCUSSION: Co-design and co-optimization: maximizing energy per joule

Keynote Address

Innovation at the Edge

Edge Trade Offs

On Device ML

Data Science

Hardware and Systems Engineering

Software Engineering

Strategy

Industry & Investment

Author:

Banu Nagasundaram

AI Product Leader

Amazon Web Services (AWS)

Read more about Keynote Address

PANEL DISCUSSION: The winning formula: combining edge strengths with cloud capabilities

Edge cloud ends the conflict of edge vs cloud but when the option is there to use the cloud, should you? This session will explore the winning combination, how experts are overcoming privacy issues and the wealth of opportunities presented by connectivity.

Innovation at the Edge

Edge Trade Offs

On Device ML

Connectivity and 5G

Hardware and Systems Engineering

Software Engineering

Strategy

Industry & Investment

Data Science

Author:

Rakshit Agrawal

Vice President of Research and Development

Camio

Author:

Charles Byers

Chief Technology Officer

Industry IoT Consortium and Valqari

Author:

Prasad Akella

Chairman and Founder

Drishti

Dr. Prasad Akella, founder and chairman of the Drishti board, is creating his third massive market category that uses technology to extend human capabilities. In the 1990s, Prasad led the General Motors team that built the world’s first collaborative robots (“cobots,” projected to be a $12B market by 2025). In the early 2000s, as cofounder of the social networking pioneer Spoke, he envisioned and helped build the first massive social graph — a category now worth trillions. Today, at Drishti, he is working to combine the cognition of AI with the flexibility of humans in factories in the form of AI-powered production. Prasad is based in Mountain View, California.

Read more about PANEL DISCUSSION: The winning formula: combining edge strengths with cloud capabilities

Panel Discussion: Edge futures: looking beyond the possible

Together, a panel of domain experts will look beyond current limitations, creating a roadmap for the future at the edge and exploring what they would want to achieve if technological barriers were removed.

Connectivity and 5G

Innovation at the Edge

Edge Trade Offs

Hardware and Systems Engineering

Software Engineering

Strategy

Industry & Investment

Author:

Gayathri Radhakrishnan

Partner

Hitachi Ventures

Gayathri is currently Partner at Hitachi Ventures. Prior to that, she was with Micron Ventures, actively investing in startups that apply AI to solve critical problems in the areas of Manufacturing, Healthcare and Automotive. She brings over 20 years of multi-disciplinary experience across product management, product marketing, corporate strategy, M&A and venture investments in large Fortune 500 companies such as Dell and Corning and in startups. She has also worked as an early stage investor at Earlybird Venture Capital, a premier European venture capital fund based in Germany. She has a Masters in EE from The Ohio State University and MBA from INSEAD in France. She is also a Kauffman Fellow - Class 16.

Author:

Sandeep Bandil

Vice President, IoT Edge Devices and Solutions

Brambles

Author:

Jordan Gosselin

Chief Engineer and Staff AI Systems Architect

Northrop Grumman

Author:

Vadim Parizher

VP of Technology

Taco Bell

Read more about Panel Discussion: Edge futures: looking beyond the possible

Healthcare

New for 2022, these discussion groups will allow attendees to delve into industry focused debate exploring the challenges and innovation driving edge AI across four specific industries.

On Device ML

Edge Trade Offs

Vision

NLP and Speech

Connectivity and 5G

Chip and Systems Design

Innovation at the Edge

Data Science

Hardware and Systems Engineering

Software Engineering

Strategy

Industry & Investment

Author:

Hooman Sedghamiz

Senior Director of AI & ML

Bayer

Hooman Sedghamiz is Director of AI & ML at Bayer. He has lead algorithm development and generated valuable insights to improve medical products ranging from implantable, wearable medical and imaging devices to bioinformatics and pharmaceutical products for a variety of multinational medical companies.

He has lead projects, data science teams and developed algorithms for closed loop active medical implants (e.g. Pacemakers, cochlear and retinal implants) as well as advanced computational biology to study the time evolution of cellular networks associated with cancer , depression and other illnesses.

His experience in healthcare also extends to image processing for Computer Tomography (CT), iX-Ray (Interventional X-Ray) as well as signal processing of physiological signals such as ECG, EMG, EEG and ACC.

Recently, his team has been working on cutting edge natural language processing and developed cutting edge models to address the healthcare challenges dealing with textual data.

Read more about Healthcare

Industrial & Manufacturing

New for 2022, these discussion groups will allow attendees to delve into industry focused debate exploring the challenges and innovation driving edge AI across four specific industries.

On Device ML

Edge Trade Offs

Vision

NLP and Speech

Connectivity and 5G

Chip and Systems Design

Innovation at the Edge

Data Science

Hardware and Systems Engineering

Software Engineering

Strategy

Industry & Investment

Author:

Christopher Nichols

Director, IT/OT Resiliency & Support

Stanley Black & Decker

Christopher Nichols is Director of IT/OT Resiliency & Support at Stanley Black & Decker, Inc. He has been employed by the Fortune 500 manufacturer, Stanley Black & Decker for over seven years, and is currently Director of IT/OT Resiliency & Support. In this role, he is responsible for deploying Level 2/3 system architectures and connecting all other level systems to level2/3, while supporting them. Additionally, he manages remediation with Edge components, and deploys servers and connectivity for all OT-Related systems, plus Level 1 support of all OT-Related SW applications.

Author:

Dominic Pajak

VP Developers

Ready Robotics

Dominic Pajak started as an engineer in Arm's CPU group in Cambridge, UK. He went on to launch energy-efficient processors that now ship in billions of electronic devices every year. Dominic has consulted with major OEMs in the automotive, consumer and industrial segments but is a big believer in making embedded technology accessible to everyone. Later at Arduino he led a collaboration with Google to launch the first developer-friendly Tiny ML library. Today as VP Developers at Ready Robotics he is working on connecting software innovators with industry-proven robotics hardware. Dominic holds a PhD in Computer Science from University of Leeds, UK.

Author:

Jake Hillard

CEO and Co-Founder

Red Leader Tech

Jake Hillard is an expert in Signal Processing and has launched multiple laser satellite missions to space. He is a Peter Theil Fellow and currently the CEO and Co-Founder of Red Leader Technologies

Author:

Sven Brunner

CEO and Founder

holo|one

Sven is founder and CEO of holo|one, a leading Mixed Reality software company. holo|one offers a standardized Mixed Reality collaboration platform targeting the high-value enterprise use cases – remote assistance, workflow guidance and live collaboration. Their solution has been adopted across the globe by Fortune 500 companies like Micron, Lenovo and the Renault Group, along with garnering several awards and accolades.

Before founding holo|one, Sven worked as a software engineer and product manager at Disney Research, ETH Zurich, and his own consulting firm. Sven holds a master’s degree in informatics from the University of Zurich and is a recipient of the Swiss National Bank’s Economics Award.

Author:

Anthony Valle

NALA Presales Senior Engineer

ATOS

Anthony Valle is a Senior Pre-Sales Engineer for North America and Latin America at Ipsotek, an Atos company. Anthony has over 20 years of experience in IT and security technology solutions for the rapidly growing tech-based world. He works closely with clients in developing solutions for AI at the Edge, utilizing a patented Scenario-Based Rule Engine (SBRE), a powerful tool to precisely define behaviors of interest as they would unfold in the real-world dynamic and complex environment.

Prior to joining Atos, he performed first Sales Engineering and later Application engineering roles for Avigilon, one of the world's largest security manufacturers. Throughout his career, he has held key management positions within the industry and sought many certifications to further his career in security technology.

Read more about Industrial & Manufacturing

Automotive

New for 2022, these discussion groups will allow attendees to delve into industry focused debate exploring the challenges and innovation driving edge AI across four specific industries.

On Device ML

Edge Trade Offs

Vision

NLP and Speech

Connectivity and 5G

Chip and Systems Design

Innovation at the Edge

Data Science

Hardware and Systems Engineering

Software Engineering

Strategy

Industry & Investment

Author:

Roger Berg

Vice President, North American Research and Development

DENSO International America, Inc.

Roger Berg is Vice President of DENSO’s North American Research and Development group. His latest research interests and responsibilities include next generation connectivity, mobile edge computing, connected automated vehicles and decentralized ledger technologies.

Author:

Gaurav Singh

Product, AI

Ridecell

Gaurav is part of the Nemo product team at Ridecell, and does advanced AI-based data analytics for ADAS & AD data. He brings a strong AI and machine learning background to his role having previously worked on developing software for autonomous driving and computer vision applications. Gaurav is a Masters in Robotics from Carnegie Mellon University.

Author:

Prashant Tiwari

Former Executive Director, Software Platforms at Volkswagen

Read more about Automotive

PANEL DISCUSSION: Choosing accuracy and robustness over performance: where are the limits?

Vision

NLP and Speech

Edge Trade Offs

On Device ML

Software Engineering

Data Science

Hardware and Systems Engineering

Strategy

Speakers

Author:

Denis Gudovskiy

Senior Deep Learning Researcher

Panasonic AI Lab

Author:

Chandra Khatri

Co-Founder

Got It AI

Chandra Khatri is the Co-Founder at Got It AI, wherein, his team is building the world's first fully autonomous Conversational AI technology. Under his leadership, Got It AI is pushing the boundaries of the Conversational AI ecosystem and delivering the next generation of automation products. Prior to Got-It, Chandra was leading various kinds of applied research groups at Uber AI such as Conversational AI, Multi-modal AI, and Recommendation Systems.

Prior to Uber AI, he was leading R&D for the Alexa Prize Competition (Alexa AI) at Amazon, wherein he got the opportunity to significantly advance the field of Conversational AI, particularly Open-domain Dialog Systems, which is considered as the holy-grail of Conversational AI and is one of the open-ended problems in AI. Prior to Alexa AI, he was driving NLP, Deep Learning, and Recommendation Systems related Applied Research at eBay. He graduated from Georgia Tech with a specialization in Deep Learning in 2015 and holds an undergraduate degree from BITS Pilani, India.

His current areas of research include Artificial and General Intelligence, Reinforcement Learning, Language Understanding, Conversational AI, Multi-modal and Human-agent Interactions, and Introducing Common Sense within Artificial Agents.

Author:

David Martin

CTO and Founder

Age@Home

Author:

Mark Kurtz

Director of Machine Learning

Neural Magic

Leader of ML and engineering teams focused on the design, development, and implementation of cutting-edge technologies and products. With over 12 years of experience in software engineering and machine learning, well-practiced in building and managing teams for both closed source and open source software solutions. Currently the Director of Machine Learning at Neural Magic, Mark is focused on lowering the cost, improving the performance, and increasing the adoption of deep learning technologies through SOTA research and engineering. An active GitHub contributor, blogger, and researcher with published papers in top ML conferences.

Neural Magic

Author:

Bernease Herman

Senior Data Scientist

WhyLabs.ai

Bernease Herman is a Sr. Data Scientist at WhyLabs, the AI Observability company, and a research scientist at the University of Washington eScience Institute. At WhyLabs, she is building model and data monitoring solutions using approximate statistics techniques. Earlier in her career, Bernease built ML-driven solutions for inventory planning at Amazon and conducted quantitative research at Morgan Stanley. Her academic research focuses on machine learning evaluation and interpretability with specialty on synthetic data and societal implications. Bernease serves as faculty for the University of Washington Master’s Program in Data Science program and as chair of the Rigorous Evaluation for AI Systems (REAIS) workshop series. She has published work in top machine learning conferences and workshops such as NeurIPS, ICLR, and FAccT. She is a PhD student at the University of Washington and holds a Bachelor’s degree in mathematics and statistics from the University of Michigan.

Moderator

Author:

Jack Kang

Senior Vice President, Business Development, Customer Experience, Corporate Marketing

SiFive

Read more about PANEL DISCUSSION: Choosing accuracy and robustness over performance: where are the limits?

VISION TRACK: Partitioning Machine Vision AI Workloads Across the Cloud – Edge – Intelligent Device Hierarchy

Machine vision workloads are complex, and their performance requirements often present challenges in areas like latency, security, energy use and reliability. Hierarchal partitioning of those workloads often makes sense, where the machine vision software is split into multiple stages (for example, contrast enhancement, feature extraction, object recognition, threat detection), which are run at different layers of the [intelligent camera -> edge node -> MEC -> cloud] hierarchy.

This talk will introduce the hierarchal cloud - edge architecture, and discuss the properties and capabilities of its many layers. It will propose an example segmentation of machine vision algorithms, and investigate the tradeoffs of how we can map them onto the various layers of processing available in the hierarchy. Finally, it will look at the dual flows of model training and inference for AI applications, and discuss which portions of those flows make sense in different edge layers, and how they can be secured, orchestrated and managed.

Vision

Edge Trade Offs

Software Engineering

Hardware and Systems Engineering

Data Science

Strategy

Author:

Charles Byers

Chief Technology Officer

Industry IoT Consortium and Valqari

Read more about VISION TRACK: Partitioning Machine Vision AI Workloads Across the Cloud – Edge – Intelligent Device Hierarchy