BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//UM//UM*Events//EN
CALSCALE:GREGORIAN
BEGIN:VTIMEZONE
TZID:America/Detroit
TZURL:http://tzurl.org/zoneinfo/America/Detroit
X-LIC-LOCATION:America/Detroit
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:20070311T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:20071104T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260112T153641
DTSTART;TZID=America/Detroit:20260128T160000
DTEND;TZID=America/Detroit:20260128T170000
SUMMARY:Workshop / Seminar:Convergence Analysis of Discrete Sampling in Continuous-Time Reinforcement Learning and High-Dimensional Numerical Integration
DESCRIPTION:Stochastic policies (also known as relaxed controls) are widely used in continuous-time Reinforcement Learning (RL) algorithms. However\, a critical disconnect remains between theory and practice. The theoretical aggregated dynamics\, driven by averaged coefficients\, provide a convenient basis for deriving RL algorithms but cannot be directly implemented. Physical execution requires the agent to sample concrete actions from the policy. Since continuously sampling independent actions poses significant mathematical and computational challenges\, practical implementation must rely on discrete sampling. Yet\, for general diffusion processes\, the accuracy of such discretely sampled dynamics has lacked rigorous theoretical justification.\n\nIn this talk\, I will bridge this gap by introducing and rigorously analyzing a policy execution framework that samples actions from a stochastic policy at discrete time points and implements them as piecewise constant controls. We prove that as the sampling mesh size tends to zero\, the controlled state process converges weakly to the dynamics with coefficients aggregated according to the stochastic policy. We explicitly quantify the convergence rate based on the regularity of the coefficients and establish an optimal first-order convergence rate for sufficiently regular coefficients. Additionally\, we prove a 1/2-order weak convergence rate that holds uniformly over the sampling noise with high probability\, and establish a 1/2-order pathwise convergence for each realization of the system noise in the absence of volatility control. Building on these results\, we analyze the bias and variance of various policy evaluation and policy gradient estimators based on discrete-time observations. Our results provide theoretical justification for the exploratory stochastic control framework in [H. Wang\, T. Zariphopoulou\, and X.Y. Zhou\, J. Mach. Learn. Res.\, 21 (2020)\, pp. 1-34].\n\nFinally\, I will also briefly discuss my research on Quasi-Monte Carlo sampling methods for efficient computation in high-dimensional numerical integration.
UID:143763-21893983@events.umich.edu
URL:https://events.umich.edu/event/143763
CLASS:PUBLIC
STATUS:CONFIRMED
CATEGORIES:Mathematics
LOCATION:Off Campus Location
CONTACT:
END:VEVENT
BEGIN:VEVENT
DTSTAMP:20260115T181513
DTSTART;TZID=America/Detroit:20260128T160000
DTEND;TZID=America/Detroit:20260128T180000
SUMMARY:Workshop / Seminar:Make Your Own Flyer Workshops
DESCRIPTION:*No prior experience required\, all supplies will be provided\"\n\nAre you curious about DIY art and culture? These art-making drop in workshops offer a hands-on way to engage with the ephemera of underground and independent music scenes featured in the Flyover Country exhibition. You will have the opportunity to make the poster for a dream show with a lineup of your favorite bands\, draw a protest flyer\, put together a collage art piece\, or even design a flyer for a real upcoming event you're planning.\nA button maker will be available for anyone who wants to turn a design into a punk-inspired button pin\, all participants will receive a free handout to take home with them.\n\nJanuary 28th workshop participants are also welcome to tape their finished flyers to a shared display wall in the gallery if they choose\, becoming part of the exhibit while it’s on view.  \n\nFree coffee\, tea\, and treats will be provided at both workshops on a first-come\, first-served basis.
UID:143980-21894368@events.umich.edu
URL:https://events.umich.edu/event/143980
CLASS:PUBLIC
STATUS:CONFIRMED
CATEGORIES:Art,Arts For All,Culture,In Person,Workshop
LOCATION:Shapiro Library - Shapiro Gallery, Third Floor, Room 3160
CONTACT:
END:VEVENT
END:VCALENDAR