The 2nd On-Device Intelligence Virtual Workshop | 9 April 2021

Held in conjunction with MLSys 2021

Ubiquitous on-device artificial intelligence (AI) is the next step in transforming the myriad of mobile computing devices in our everyday lives into a new class of truly “smart” devices, capable of constantly observing, learning, and adapting to their environment. These intelligent devices can make our lives safer and the world around us more energy efficient.

The second On-Device Intelligence Workshop aims to advance the state of the art by bringing together researchers and practitioners to discuss the key problems and share new research results and practical tutorial material.

Understand the Challenges and Solutions

Community building and networking is vital to facilitate future collaboration across experts in algorithms, software and hardware engineering domains. What can your expertise bring to the table?


In keynote presentations, talks, and posters, experts discuss four key challenges:

How do we design, train, and optimize machine learning (ML) models tailored to fit a plethora of edge devices with constrained compute, storage and energy budgets?

How do we implement distributed ML systems on-device, that are able to collaboratively exploit data, while preserving privacy?

How should mobile computing hardware evolve to support the increasing prevalence of on-device AI workloads?

How can industry and academia collaboratively develop standards and benchmarks to stimulate the development of an on-device AI research ecosystem?

Look Forward to Talks from Experts

Talk Presenter
Keynote: Putting AI on Diet: TinyML and Efficient Deep Learning Song Han, MIT
Efficient ML on the Edge with Apache TVM Thierry Moreau, OctoML
Customizing Federated Learning to the Edge Device Venkatesh Saligrama, Boston University
Enabling standardized behavior for ML operations with TOSA

Eric Kunze, Arm

On-Device NLP at Facebook Ahmed Aly, Facebook Reality Labs
Lightning Tutorial: TinyML model design Igor Federov, Arm Research
Lightning Tutorial: TinyML HW evaluation Colby Banbury, Harvard University


Register for MLSys 2021 to view the full schedule, or get in touch with Paul Whatmough.

Full Workshop Schedule

Organizing Committee

Have a question?

Contact Paul Whatmough.

Contact Paul