BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Asia/Tokyo
X-LIC-LOCATION:Asia/Tokyo
BEGIN:STANDARD
TZOFFSETFROM:+0900
TZOFFSETTO:+0900
TZNAME:JST
DTSTART:18871231T000000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20250110T023302Z
LOCATION:G610\, G Block\, Level 6
DTSTART;TZID=Asia/Tokyo:20241203T130000
DTEND;TZID=Asia/Tokyo:20241203T144500
UID:siggraphasia_SIGGRAPH Asia 2024_sess210_crs_119@linklings.com
SUMMARY:MCMC: Bridging Rendering, Optimization and Generative AI
DESCRIPTION:Courses\n\nGurprit Singh and Gurprit Singh (Max Planck Institu
 te for Informatics) and Wenzel Jakob (EPFL)\n\nGenerative artificial intel
 ligence (AI) has made unprecedented advances in vision language models ove
 r the past two years. These advances are largely due to diffusion-based ge
 nerative models, which are very stable and simple to train.  These diffusi
 on models are tasked to learn the underlying unknown distribution of the t
 raining data samples. During the generative process, new samples (images) 
 are generated from this unknown high-dimensional distribution. Markov Chai
 n Monte Carlo (MCMC) methods are particularly effective in drawing samples
  from complex, high-dimensional distributions. This makes MCMC methods an 
 integral component for both the training and sampling phases of these mode
 ls, ensuring accurate sample generation.\n    \n    Gradient-based optimiz
 ation is at the core of modern generative models. The update step during t
 he optimization forms a Markov chain where the new update depends only on 
 the current state. This allows exploration of the parameter space in a mem
 oryless manner, thus combining the benefits of gradient-based optimization
  and MCMC sampling. MCMC methods have shown an equally important role in p
 hysically based rendering where complex light paths are otherwise quite ch
 allenging to sample from simple importance sampling techniques. \n    \n  
   A lot of research is dedicated towards bringing physical realism to samp
 les (images) generated from diffusion-based generative models in a data-dr
 iven manner, however, a unified framework connecting these techniques is s
 till missing. In this course, we take the first steps toward understanding
  each of these components and exploring how MCMC could potentially serve a
 s a bridge, linking these closely related areas of research. Our course ai
 ms to provide necessary theoretical and practical tools to guide students,
  researchers and practitioners towards the common goal of generative physi
 cally based rendering.\n\nRegistration Category: Full Access, Full Access 
 Supporter\n\nLanguage Format: English Language
URL:https://asia.siggraph.org/2024/program/?id=crs_119&sess=sess210
END:VEVENT
END:VCALENDAR