BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Biomedical Mathematics Group - ECPv6.15.20//NONSGML v1.0//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
X-WR-CALNAME:Biomedical Mathematics Group
X-ORIGINAL-URL:https://www.ibs.re.kr/bimag
X-WR-CALDESC:Events for Biomedical Mathematics Group
REFRESH-INTERVAL;VALUE=DURATION:PT1H
X-Robots-Tag:noindex
X-PUBLISHED-TTL:PT1H
BEGIN:VTIMEZONE
TZID:Asia/Seoul
BEGIN:STANDARD
TZOFFSETFROM:+0900
TZOFFSETTO:+0900
TZNAME:KST
DTSTART:20230101T000000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTART;TZID=Asia/Seoul:20241227T100000
DTEND;TZID=Asia/Seoul:20241227T120000
DTSTAMP:20260430T165626
CREATED:20241022T001840Z
LAST-MODIFIED:20241226T235355Z
UID:10197-1735293600-1735300800@www.ibs.re.kr
SUMMARY:Diffusion Posterior Sampling for Linear Inverse Problem Solving- A Filtering Perspective - U Jin Choi
DESCRIPTION:In this talk\, we discuss the paper : “Diffusion Posterior Sampling for Linear Inverse Problem Solving- A Filtering Perspective” by Z. Dou& Y. Song \n\n\nDiffusion models have achieved tremendous success in generating high-dimensional data like images\, videos and audio. These models provide powerful data priors that can solve linear inverse problems in zero shot through Bayesian posterior sampling. However\, exact posterior sampling for diffusion models is intractable. Current solutions often hinge on approximations that are either computationally expensive or lack strong theoretical guarantees. In this work\, we introduce an efficient diffusion sampling algorithm for linear inverse problems that is guaranteed to be asymptotically accurate. We reveal a link between Bayesian posterior sampling and Bayesian filtering in diffusion models\, proving the former as a specific instance of the latter. Our method\, termed filtering posterior sampling\, leverages sequential Monte Carlo methods to solve the corresponding filtering problem. It seamlessly integrates with all Markovian diffusion samplers\, requires no model re-training\, and guarantees accurate samples from the Bayesian posterior as particle counts rise. Empirical tests demonstrate that our method generates better or comparable results than leading zero-shot diffusion posterior samplers on tasks like image inpainting\, super-resolution\, and deblurring.
URL:https://www.ibs.re.kr/bimag/event/diffusion-posterior-sampling-for-linear-inverse-problem-solving-a-filtering-perspective-u-jin-choi/
LOCATION:B232 Seminar Room\, IBS\, 55 Expo-ro Yuseong-gu\, Daejeon\, Daejeon\, 34126\, Korea\, Republic of
CATEGORIES:Biomedical Mathematics Seminar
ORGANIZER;CN="Jae Kyoung Kim":MAILTO:jaekkim@kaist.ac.kr
END:VEVENT
END:VCALENDAR