Monday, September 30, 2019
12:00 PM
1400 Wegmans Hall
Asif Salekin
Syracuse University
Adaptive Machine Learning for IoT Applications
With the increasing interest in smart cars, smart homes, smart health-care, and smart cities, we are moving toward a smart world. Supported by rapid innovations in machine learning, signal processing, computing, and wearable-systems, the concept of smart and connected sensing is redesigning almost every aspect of our lives. Innovating novel, low cost and noninvasive sensing techniques to model/identify an individualís behavioral or physiological states is emerging as one of the core research areas. Yet, the accuracy of detecting or sensing human events is often not adequate to have any practical use. This is due to the lack of adaptability of the current state-of-the-art machine learning and data analytics techniques, with the characteristics and constraints of real-word sensing applications. This presentation aims to demonstrate sensing and data analytics approaches that address the challenges of real-world applications, such as, the uncertainties in physical world sensing, human factors, such as, the user-context and mobility, the limitation of current technologies, and resource constraints of the sensing data and computation platform. I will showcase these approaches via two of my recent works: detecting mental disorder from weakly labelled data and DER: distant emotion recognition. In later part of the presentation, I will present a brief overview of my current works.

Asif Salekin is an assistant professor in the Department of Electrical Engineering and Computer Science at the Syracuse University. His research takes a multi-disciplinary approach to develop novel and practical human event sensing technologies that capture observable low-level physical signals from human bodies and surrounding environments and employ new machine learning, signal processing and natural language processing techniques, to rectify the existing sensing technologies. Asifís research has been well-accepted by a diverse set of communities. He has published his works at the leading conferences in the fields of ubiquitous computing, connected health, and wireless sensor networks. In 2018, he received the Graduate Student Award for Outstanding Research from UVA, Department of Computer Science. His work, AsthmaGuide, got nominated for the best paper award in the Wireless Health 2016 conference. Additionally, he has gained the experience of working in two leading industrial research labs (Nokia Bell Labs and Bosch Research) as a research intern.