BEGIN:VCALENDAR
VERSION:2.0
PRODID:IEEE vTools.Events//EN
CALSCALE:GREGORIAN
BEGIN:VTIMEZONE
TZID:America/New_York
BEGIN:DAYLIGHT
DTSTART:20260308T030000
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
RRULE:FREQ=YEARLY;BYDAY=2SU;BYMONTH=3
TZNAME:EDT
END:DAYLIGHT
BEGIN:STANDARD
DTSTART:20251102T010000
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
RRULE:FREQ=YEARLY;BYDAY=1SU;BYMONTH=11
TZNAME:EST
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20251120T164357Z
UID:CB2226F9-5D0A-46DB-8202-EE2120959E30
DTSTART;TZID=America/New_York:20251118T190000
DTEND;TZID=America/New_York:20251118T200000
DESCRIPTION:The robust learning of dynamical systems is crucial for safety-
 critical applications\, such as power systems and autonomous systems. The 
 control theory has a rich literature on system identification and optimal 
 control in the case when the system is subject to non-adversarial and most
 ly Gaussian disturbance. However\, there is a pressing need to develop lea
 rning and control techniques for systems whose inputs are under adversaria
 l attacks\, meaning that the system operates in a hostile environment. The
  main challenge is that the classic results relying on closed-form solutio
 ns for least-square estimators and LQR/LQG are no longer valid\, and it is
  essential to design non-smooth estimators and controllers with no closed-
 form solutions in presence of adversarial attacks. In this talk\, we discu
 ss the recent advances in the area and focus on the problem of learning an
  unknown nonlinear dynamical system subject to adversarial disturbance/inp
 ut. We develop a non-smooth estimator and show that the correct dynamics o
 f the system can be learned in finite time no matter how severe the attack
  is as long as the learning period is longer than some threshold. We then 
 study optimal control for systems in hostile environments.\n\nSpeaker(s): 
 Dr.Lavaei\, \n\nVirtual: https://events.vtools.ieee.org/m/502766
LOCATION:Virtual: https://events.vtools.ieee.org/m/502766
ORGANIZER:joseph.palko@ieee.org
SEQUENCE:36
SUMMARY:OPTIMAL CONTROL AND LEARNING FOR DYNAMICAL SYSTEMS UNDER ADVERSARIA
 L ATTACKS
URL;VALUE=URI:https://events.vtools.ieee.org/m/502766
X-ALT-DESC:Description: <br /><p><span lang="IT" style="font-size: 11.0pt\;
  font-family: 'Aptos'\,sans-serif\; mso-ascii-theme-font: minor-latin\; ms
 o-fareast-font-family: Aptos\; mso-fareast-theme-font: minor-latin\; mso-h
 ansi-theme-font: minor-latin\; mso-bidi-font-family: 'Times New Roman'\; m
 so-bidi-theme-font: minor-bidi\; mso-ansi-language: IT\; mso-fareast-langu
 age: EN-US\; mso-bidi-language: AR-SA\;">The robust learning of dynamical 
 systems is crucial for safety-critical applications\, such as power system
 s and autonomous systems. The control theory has a rich literature on syst
 em identification and optimal control in the case when the system is subje
 ct to non-adversarial and mostly Gaussian disturbance. However\, there is 
 a pressing need to develop learning and control techniques for systems who
 se inputs are under adversarial attacks\, meaning that the system operates
  in a hostile environment. The main challenge is that the classic results 
 relying on closed-form solutions for least-square estimators and LQR/LQG a
 re no longer valid\, and it is essential to design non-smooth estimators a
 nd controllers with no closed-form solutions in presence of adversarial at
 tacks. In this talk\, we discuss the recent advances in the area and focus
  on the problem of learning an unknown nonlinear dynamical system subject 
 to adversarial disturbance/input. We develop a non-smooth estimator and sh
 ow that the correct dynamics of the system can be learned in finite time n
 o matter how severe the attack is as long as the learning period is longer
  than some threshold. We then study optimal control for systems in hostile
  environments.</span></p>
END:VEVENT
END:VCALENDAR