https://murray.cds.caltech.edu/api.php?action=feedcontributions&feedformat=atom&user=Anima 2026-04-07T21:45:51Z User contributions MediaWiki 1.41.5 https://murray.cds.caltech.edu/index.php?title=Sarah_Dean,_11-12_Feb_2020&diff=23356 2020-02-05T17:32:25Z

<p>Anima: /* Tuesday (11 Feb) */</p> <hr /> <div>Sarah Dean, a PhD student working with Ben Recht, will visit Caltech on 11-12 Feb 2020. If you would like to meet with her, please sign up for a slot below (using your IMSS credentials to log in). Please make sure to put the location where she should meet you.<br /> <br /> === Schedule ===<br /> <br /> {| width=100%<br /> |- valign=top <br /> | width=50% |<br /> ==== Tuesday (11 Feb) ====<br /> * 11:40 am arrival in BUR<br /> * ~12:15 pm: arrival on campus<br /> * 12:15 pm: John Doyle<br /> * 1:00 pm: Quick lunch (TBD)<br /> * 1:30 pm: Richard Murray (109 Steele Lab)<br /> * 2:00 pm: Open<br /> * 2:45 pm: Open<br /> * 3:30 pm: Anima Anandkumar (30 min)<br /> * 4:00 pm: Seminar<br /> * 5:00 pm: Open (30 min)<br /> * 6:00 pm: Dinner with Richard + grad students, postdocs<br /> | width=50% |<br /> <br /> ==== Wednesday (12 Feb) ====<br /> * 8:45 am: Open<br /> * 9:30 am: Open<br /> * 10:15 am: Open<br /> * 11:00 am: Ludwig Schmidt seminar<br /> * 12:00 pm: Lunch with faculty or grad students<br /> * 1:15 pm: Yisong Yue (303 Annenberg)<br /> * 2:00 pm: Aaron Ames (266 Gates-Thomas)<br /> * 2:30 pm: Meet with Ames' Students (121 Gates Thomas)<br /> * 3:00 pm: CDS tea<br /> * 3:30 pm: Open (30 min)<br /> * 4:00 pm: Open (30 min)<br /> * 4:30 pm: Wrap up meeting with Richard (109 Steele Lab)<br /> * 4:45 pm: Depart campus<br /> * 6:40 pm departure from BUR<br /> |}<br /> <br /> === Seminar ===<br /> <br /> Safe and Robust Perception-Based Control&lt;br&gt;<br /> Sarah Dean, UC Berkeley<br /> <br /> Tue, 11 February, 4 pm&lt;br&gt;<br /> 105 Annenberg<br /> <br /> Machine learning provides a promising path to distill information from high dimensional sensors like cameras -- a fact that often serves as motivation for merging learning with control. This talk aims to provide rigorous guarantees for systems with such learned perception components in closed-loop. Our approach is comprised of characterizing uncertainty in perception and then designing a robust controller to account for these errors. We use a framework which handles uncertainties in an explicit way, allowing us to provide performance guarantees and illustrate how trade-offs arise from limitations of the training data. Throughout, I will motivate this work with the example of autonomous vehicles, including both simulated experiments and an implementation on a 1/10 scale autonomous car. Joint work with Aurelia Guy, Nikolai Matni, Ben Recht, Rohan Sinha, and Vickie Ye.</div>

Anima