‘Architecture and Systems’
‘Vision, Networks, and Control Theory’
‘Natural Language Processing (NLP)’

Presented by:
Mike Ferdman
Romeil Sandhu
Niranjan Balasubramanian
Ritwik Banerjee

Geometry and Control Towards Vision, Networks, and Autonomy

Abstract:

Over the past twenty years, we have witnessed a dramatic increase in the availability of time-varying information acquired through varying systems. The two most common interpretable forms of such data irrespective of applications are images and networks. To this end, we have inundated with varying machine learning algorithms (e.g., deep learning) in order to make sense of such data with ambitious claims of an “artificial intelligence takeover.” However, while such techniques have aided our understanding in robotic vision, autonomy, and varying applications in medicine, there still exists an “information gap” to combat the “unknown unknown” for which we often require a human to oversee the operation of such tasks. Unfortunately, machine learning alone will not be able to handle complex real-world situations where one needs to eliminate and not mitigate probability of extreme events (i.e., loss of life). To this end, this talk will focus on the how geometry and control is a foundational element in vision, robotics, and broader data science.

This talk is intended for incoming graduate students who are interested in working in the fields of computer vision, network/graph theory, robotics, autonomy. Most importantly, this talk is for students who are interested in the above fields yet also possess a desire towards the understanding of how the mathematical mechanics in geometry, control, and statistics can be applied towards such problems and beyond.

Bio:

Romeil Sandhu is currently an Assistant Professor at Stony Brook University with appointments in Computer Science, Biomedical Informatics, and Applied Mathematics & Statistics Departments and is the recipient of the 2018 AFOSR YIP Award for work on interactive feedback control for autonomous systems and 2018 NSF CAREER Award for work on geometric optimization of time-varying networks. Romeil first received his B.S. and M.S. degrees from the Georgia Institute of Technology in Electrical Engineering in 2006 and 2009, respectively. Then, under the direction of Professor Allen Tannenbaum, he completed his Ph.D. in 2011 in control-based vision and learning. Prior to his academic position at Stony Brook, he formed a startup providing government services leading to a successful exit. His current research interest focuses on formally developing theory to bridge the information gap to combat risk complexities associated with the “unknown unknown” with a particular interest in the fusion of autonomous systems towards “non-expert” operators who possesses expert knowledge.