IEEE Seoul Section Sensors Council Chapter
International Workshop 2021 (Virtual,Free)
(ai-security.org)

Program Schedule

16 December 2021

Time Program Speaker
09:00 - 10:00
(KST)
Recent Trends in Biometric: Open Set Presentation Attack Detection Prof. Kevin W. Bowyer
10:00 - 11:00
(KST)
Trends for Wearable and Medical Devices Prof. Subhas Mukhopadhyay

17 December 2021

Time Program Speaker
09:00 - 10:00
(KST)
Quantifying Uncertainty in Machine Learning Based Sensing Prof. Shervin Shirmohammadi
10:00 - 11:00
(KST)
Efficient Deep Learning at Scale: Hardware and Software Prof. Yiran Chen
Synopses

16 December 2021

Recent Trends in Biometric: Open Set Presentation Attack Detection
Speaker:Prof. Kevin W. Bowyer, University of Notre Dame, USA
Biometric presentation attack detection is made especially difficult because of its open-set nature in the real world. The current highest-accuracy algorithms for iris presentation attack detection use deep learning approaches.
We show a novel approach to achieve greater accuracy in deep learning from limited training data using human-aided saliency maps.

Related material: https://arxiv.org/pdf/2105.03492.pdf
Trends for Wearable and Medical Devices
Speaker: Prof. Subhas Mukhopadhyay, Macquarie University, Australia
An increase in world population along with a significant aging portion is forcing rapid rises in healthcare costs. The healthcare system is going through a transformation in which continuous monitoring of inhabitants is possible even without hospitalization. The advancement of sensing technologies, embedded systems, wireless communication technologies, nano-technologies, and miniaturization makes it possible to develop smart medical systems to monitor activities of human beings continuously.
Wearable sensors monitor physiological parameters continuously along with detect other symptoms such as any abnormal and/or unforeseen situations which need immediate attention. Therefore, necessary help can be provided in times of dire need. This seminar reviews the latest reported systems and the trends on wearable and medical devices to monitor activities of humans and issues to be addressed to tackle the challenges.

17 December 2021

Quantifying Uncertainty in Machine Learning Based Sensing
Speaker:Prof. Shervin Shirmohammadi, Ottawa University, Canada
Like any science and engineering field, Instrumentation and Measurement (I&M) including sensors are currently experiencing the impact of the recent rise of Applied AI and in particular Machine Learning (ML). In fact the relationship between I&M and ML has reached new levels: sensors are used to measure and collect data, which is used to train an ML model, which is then used in a sensor system or application. Uncertainty is accumulated at every stage, and quantifying it is crucial. But I&M and ML use terminology that sometimes sound or look similar, though they might only have a marginal relationship or even be false friends. Therefore, understanding the terminology used by both communities is of crucial importance to understand the influences of ML and I&M in each other.
In this talk, we will give an overview of ML’s contribution to a sensor’s measurement error, and how to avoid confusion with the said terminology in order to better understand the application of ML in sensor measurements. We then use that understanding and terminology to show how to quantify the uncertainty introduced by ML, specifically Deep learning (DL), in DL-based sensor systems and applications.
Efficient Deep Learning at Scale: Hardware and Software
Speaker:Prof. Yiran Chen, Duke University, USA
The rapid growth of modern neural network models’ scale generates ever-increasing demands for high computing power of Artificial Intelligence (AI) systems. Many specialized computing devices have been also deployed in the AI systems, forming a truly application-driven heterogeneous computing platform. This talk discusses the importance of hardware/software co-design in the development of AI computing systems.
We first use resistive memory based Neural Network (NN) accelerators to illustrate the design philosophy of heterogeneous AI computing systems, and then present several hardware-friendly neural network model compression techniques.
We also extend our discussions to distributed systems and briefly introduce the automation of the co-design flow, e.g., neural architecture search. A research roadmap of our relevant research is given at the end of the talk.