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:20261101T010000
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
RRULE:FREQ=YEARLY;BYDAY=1SU;BYMONTH=11
TZNAME:EST
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260319T010128Z
UID:CF070BD4-9B6A-4F57-AC7F-2AE5E7407457
DTSTART;TZID=America/New_York:20260318T163000
DTEND;TZID=America/New_York:20260318T180000
DESCRIPTION:I will present safe learning methods for autonomy that combine 
 Bayesian modeling with control barrier formulations to enable real-time sa
 fety and online adaptation. I will then introduce DAREK\, a distance-aware
  uncertainty model that offers a computationally efficient alternative to 
 Gaussian-process-based approaches. Next\, I will move from control to sens
 ing and inverse problems. I will discuss how invertible architectures enab
 le efficient posterior inference\, and how symbolic invertibility can furt
 her improve interpretability. I will illustrate these ideas on an inverse 
 problem in ocean-acoustic sensing.\n\nCo-sponsored by: IEEE CIS and CS Sch
 enectady Chapter\n\nSpeaker(s): Mohammad Javad\n\nVirtual: https://events.
 vtools.ieee.org/m/546062
LOCATION:Virtual: https://events.vtools.ieee.org/m/546062
ORGANIZER:lkmestha@ieee.org;daniel.jeevaguntala@ieee.org
SEQUENCE:56
SUMMARY:IEEE CIS & CS Schenectady Chapters Technical Lecture on "Efficient 
 Learning for Control and Sensing"
URL;VALUE=URI:https://events.vtools.ieee.org/m/546062
X-ALT-DESC:Description: <br /><p>I will present safe learning methods for a
 utonomy that combine Bayesian modeling with control barrier formulations t
 o enable real-time safety and online adaptation. I will then introduce DAR
 EK\, a distance-aware uncertainty model that offers a computationally effi
 cient alternative to Gaussian-process-based approaches. Next\, I will move
  from control to sensing and inverse problems. I will discuss how invertib
 le architectures enable efficient posterior inference\, and how symbolic i
 nvertibility can further improve interpretability. I will illustrate these
  ideas on an inverse problem in ocean-acoustic sensing.&nbsp\;</p>
END:VEVENT
END:VCALENDAR