8:45 – 9:00 Introduction and Opening Remarks
9:00 – 10:00 Keynote Talk
Machine Learning Systems: On-device AI and Beyond
Dr. Nicholas D. Lane (University of Oxford, UK)
Abstract: In just a few short years, breakthroughs from the field of deep learning have transformed how computational models perform a wide-variety of tasks such as recognizing a face, driving a car or translate a language. Unfortunately, deep models and algorithms typically exert severe demands on local device resources and this conventionally limits their adoption within mobile and embedded platforms. Because sensor perception and reasoning are so fundamental to this class of computation, I believe the evolution of devices like phones, wearables and things will be crippled until we reach a point where current -- and future -- deep learning innovations can be simply and efficiently integrated into these systems.
In this talk, I will describe our progress towards developing general-purpose support for deep learning on resource-constrained mobile and embedded devices. Primarily, this requires a radical reduction in the resources (viz. energy, memory and computation) consumed by these models -- especially at inference time. I will highlight various, largely complementary, approaches we have invented to achieve this goal including: binary “on-the-fly” networks, sparse layer representations, dynamic forms of compression, and scheduling partitioned model architectures. Collectively, these techniques rethink how deep learning algorithms can execute not only to better cope with mobile and embedded device conditions; but also to increase the utilization of commodity processors (e.g., DSPs, GPUs, CPUs) -- as well as emerging purpose-built deep learning accelerators.
10:30 – 12:00 Session 1: Gestures and hand Motions
1.1 Firearm Detection Using Wrist Worn Tri-Axis Accelerometer Signals
Md Abdullah Al Hafiz Khan, David Welsh and Nirmalya Roy (University of Maryland Baltimore County, USA)
1.2 URQUELL: Using wrist-based gestural interaction to discover POIs in urban environments
Andreas Komninos, Stylianos Sarakinos, Lydia Marini and John Garofalakis (University of Patras, Greece)
1.3 Detecting Distracted Driving Using a Wrist-Worn Wearable
Bharti Goel, Arup Kanti Dey, Pratool Bharti, Kaoutar Ben Ahmed and Sriram Chellappan (University of South Florida, USA)
1.4 Towards a Practical Pedestrian Distraction Detection Framework using Wearables
Nisha Vinayaga-Sureshkanth (The University of Texas at San Antonio, USA); Anindya Maiti (Wichita State University, USA); Murtuza Jadliwala (The University of Texas at San Antonio, USA); Kirsten Crager and Jibo He (Wichita State University, USA); Heena Rathore (Hiller Measurements, USA)
12:00 – 13:30 Lunch break
13:30 – 15:00 Session 2: Beyond activity recognition
2.1 A Lifelogging Platform Towards Detecting Negative Emotions in Everyday Life using Wearable Devices
Chelsea Dobbins, Stephen Fairclough and Paulo Lisboa (Liverpool John Moores University, United Kingdom (Great Britain)); Félix Navarro (Instituto de Ingeniería Universidad Autónoma de Baja California, Mexico)
2.2 Biofeedback in the Wild - A SmartWatch Approach
Nico Pilgram, Ahmed Mohamed, Matthias Kranz and Gerold Hölzl (University of Passau, Germany)
2.3 PresentPostures: A Wrist and Body Capture Approach for Augmenting Presentations
Jochen Kempfle and Kristof Van Laerhoven (University of Siegen, Germany)
2.4 Decoupling Screen Size and Gesture Size for Wrist Worn Devices
Michael Jones and Kevin Seppi (Brigham Young University, USA); Jared Forsyth (Khan Academy, USA); Zann Anderson (Brigham Young University, USA)
15:00 – 15:30 Coffee break
15:30 – 16:30 Session 3: Evaluating wrist sensing technologies
3.1 A Comprehensive Comparison of Commercial Wrist-Worn Trackers in a Young Cohort in a Lab-Environment
Salvatore Tedesco, Marco Sica and John Barton, Brendan O'Flynn, Thomas Garbay (Tyndall National Institute, Ireland)
3.2 Panel Discussion Regarding How to Establish Objective Gold Standard Measurements for Health Constructs.
Due to participants' increased willingness to wear wrist-worn sensors, the wrist remains a major contender for tackling problems related to health and medicine. Currently physical and sedentary activity are the only constructs with acceptable gold standard passive sensing measures (e.g. Actigraph measuring MVPA and Actipal measuring types of sedentary activity). We aim to begin the discussion to change that. We have put together a unique panel discussion comprising key individuals across government agencies, academia and industry surrounding "Defining Objective Gold Standard Measures for Health Outcomes." Our goal is to engage the research community in this important discussion, and perhaps identify a vision moving forward for technologist, behaviorist, engineers, and mHealth specialist.
Expert Panelists:
Dr. Claudio Bettini (University of Milan, Italy)
Claudio Bettini is full professor at the computer science department of the University of Milan, where he leads the EveryWare laboratory. He received his PhD in Computer Science from the University of Milan in 1993. He has been post-doc at IBM Research, NY, and, for more than a decade, an affiliate research professor at the Center for Secure Information Systems at George Mason University, VA. His research interests cover the areas of mobile and pervasive computing, data privacy, temporal and spatial data management, and eHealth applications. On these topics he has published extensively in international conferences and scientific journals. He has been playing major roles in the organization of international events in the Database and Pervasive Computing areas, among which TPC Chair of IEEE PERCOM 2013 (International Conference on Pervasive Computing and Communications), and general chair of IEEE PERCOM 2017 and IEEE MDM 2013 (International Conference on Mobile Data Management). He is currently a member of the steering committee of IEEE PerCom and associate editor of the Pervasive and Mobile Computing Journal and previously editor of The VLDB Journal and of the IEEE Transactions on Knowledge and Data Engineering. In 2011 he founded EveryWare Technologies, a startup developing innovative mobile apps for privacy and assistive technologies. He is a member of ACM SIGMOD and senior member of the IEEE Computer Society.
Dr. Adam Haim (NIMH - USA)
Dr. Adam Haim is the Chief of the Treatment and Preventive Intervention Research Branch within the Division of Services and Intervention Research at the National Institute of Mental Health (NIMH). Dr. Haim manages a broad portfolio of research focused on evaluating the efficacy and effectiveness of pharmacologic, psychosocial and combination interventions on mental and behavior disorders. He is also a thought leader in the development, evaluation and implementation of technology enhanced mental health interventions. Dr. Haim is a licensed clinical psychologist and earned his doctoral degree in clinical psychology from State University of New York at Albany and completed his research fellowship at the NIMH Intramural Program in the Division of Clinical Neuroendocrinology.
Dr. Andreas Lymberis (European Commission - Belgium)
Andreas Lymberis is a physicist, post-graduated with a Ph.D. in biomedical engineering and sciences (1990, Paris, France). He worked for more than 20 years as researcher/engineer and R&D manager in biomedical technology and health telematics. In 1999 he joined the European Commission (Brussels, Belgium) as a scientific officer in eHealth where he initiated R&D activities on “smart wearable health systems and biomedical clothing”. Since 2004 he is senior research program officer first in microsystems and recently in electronic components and systems. In 2016 he has been appointed as head of sector "Wearables and Bioelectronics". He is Senior IEEE Member and chaired the IEEE-EMBS Technical Committee on Wearable Biomedical Sensors and Systems (2004-2007). He published over 60 articles in journals, conference proceedings & books and he is editor of 2 books on wearable and mobile health systems.
16:30 – 17:00 Awards and closing remarks
