BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//132.216.98.100//NONSGML kigkonsult.se iCalcreator 2.20.4//
BEGIN:VEVENT
UID:20260415T223202EDT-7037FWerJO@132.216.98.100
DTSTAMP:20260416T023202Z
DESCRIPTION:Virtual Informal Systems Seminar (VISS)\n	Centre for Intelligent
  Machines (CIM) and Groupe d'Etudes et de Recherche en Analyse des Decisio
 ns (GERAD)\n	\n	Zoom Link \n	Meeting ID: 910 7928 6959        \n	Passcode: VIS
 S  \n	\n	Speaker: Necmiye Ozay\, Associate Professor\, Electrical Engineerin
 g and Computer Science\, University of Michigan\n	\n	Abstract: System identi
 fication has a long history with several well-established methods\, in par
 ticular for learning linear dynamical systems from input/output data. Whil
 e the asymptotic properties of these methods are well understood as the nu
 mber of data points goes to infinity or the noise level tends to zero\, ho
 w well their estimates in finite data regime evolve is relatively less stu
 died. This talk will mainly focus on our analysis of the robustness of the
  classical Ho-Kalman algorithm and how it translates to non-asymptotic est
 imation error bounds as a function of the number of data samples. In the s
 econd part of the talk\, I will describe a practical problem where a robot
  needs to learn safe behaviors from a limited number of demonstrations. We
  recast this problem as an inverse constraint learning problem\, similar t
 o inverse optimal control. Our experiments with several robotics problems 
 show (local) optimality can be a very strong prior in learning from demons
 trations. I will conclude the talk with some open problems and directions 
 for future research.\n	\n	Biography: Necmiye Ozay received her B.S. degree f
 rom Bogazici University\, Istanbul in 2004\, her M.S. degree from the Penn
 sylvania State University\, University Park in 2006 and her Ph.D. degree f
 rom Northeastern University\, Boston in 2010\, all in electrical engineeri
 ng. She was a postdoctoral scholar at the California Institute of Technolo
 gy\, Pasadena between 2010 and 2013. She joined the University of Michigan
 \, Ann Arbor in 2013\, where she is currently an associate professor of El
 ectrical Engineering and Computer Science. She is also a member of the Mic
 higan Robotics Institute. Dr. Ozay's research interests include hybrid dyn
 amical systems\, control\, optimization and formal methods with applicatio
 ns in cyber-physical systems\, system identification\, verification and va
 lidation\, autonomy and dynamic data analysis. Her papers received several
  awards. She received the 1938E Award and a Henry Russel Award from the Un
 iversity of Michigan for her contributions to teaching and research\, and 
 five young investigator awards\, including NSF CAREER. She is also a recip
 ient of the 2021 Antonio Ruberti Young Researcher Prize from the IEEE Cont
 rol Systems Society for her fundamental contributions to the control and i
 dentification of hybrid and cyber-physical systems.\n	 \n
DTSTART:20220318T140000Z
DTEND:20220318T150000Z
LOCATION:CA\, ZOOM
SUMMARY:Learning Models and Constraints with Limited Data
URL:https://www.mcgill.ca/cim/channels/event/learning-models-and-constraint
 s-limited-data-338443
END:VEVENT
END:VCALENDAR