Title: DDPM Score Matching and Distribution Learning Abstract: Score estimation is the backbone of score-based generative models (SGMs), especially denoising diffusion probabilistic models (DDPMs). A key result in this area shows that with accurate score estimates, SGMs can efficiently generate samples from any realistic data distribution (Chen et al., ICLR'23; Lee et al., ALT'23). However, this distribution learning result, where the learned distribution is implicitly that of the sampler's output, does not explain how score estimation relates to classical tasks of parameter and density estimation. In this talk, we will discuss a framework that reduces these two tasks to score estimation, with various implications for statistical and computational learning theory. Bio: Alkis Kalavasis is a Postdoctoral Fellow at the Institute for Foundations of Data Science (FDS) at Yale University. He earned his PhD at the National Technical University of Athens (NTUA), advised by Dimitris Fotakis and Christos Tzamos. His research lies at the intersection of machine learning, statistics, and data science, with a particular focus on learning from imperfect and missing data—spanning applications in causal inference, generative modeling, and multi-agent systems. His work has been recognized with the Best Paper Award at COLT 2025, the Best Paper Award at the DeLTA Workshop at ICLR 2025, and the 2024 Dimitris Chorafas Award for his doctoral thesis. Alkis Kalavasis is visiting Stratis Skoulakis. ________________________________________________________________________________ Microsoft Teams Har du brug for hjælp?<https://aka.ms/JoinTeamsMeeting?omkt=da-DK> Deltag i mødet nu<https://teams.microsoft.com/l/meetup-join/19%3ameeting_ZmQyZTBhOWQtMmI5Yi00ZmEwLThjN2QtMGNkYjFkYzVhYWM3%40thread.v2/0?context=%7b%22Tid%22%3a%2261fd1d36-fecb-47ca-b7d7-d0df0370a198%22%2c%22Oid%22%3a%226f10f640-b1a2-4efa-8502-b45bbb454537%22%7d> Møde-id: 346 695 106 284 4 Adgangskode: 8Xx997hz ________________________________ Ring op via telefon +45 32 73 55 36,,634962479#<tel:+4532735536,,634962479> Danmark, All locations Find et lokalt nummer<https://dialin.teams.microsoft.com/8ad9a387-92fd-4e73-9dd7-bfa4e94dd900?id=634962479> Telefonmøde-id: 634 962 479# Tilmeld dig på en enhed til videomøder Lejernøgle: aarhusuni@m.webex.com<mailto:aarhusuni@m.webex.com> Video-id: 128 650 306 0 Flere oplysninger<https://www.webex.com/msteams?confid=1286503060&tenantkey=aarhusuni&domain=m.webex.com> For arrangører: Mødeindstillinger<https://teams.microsoft.com/meetingOptions/?organizerId=6f10f640-b1a2-4efa-8502-b45bbb454537&tenantId=61fd1d36-fecb-47ca-b7d7-d0df0370a198&threadId=19_meeting_ZmQyZTBhOWQtMmI5Yi00ZmEwLThjN2QtMGNkYjFkYzVhYWM3@thread.v2&messageId=0&language=da-DK> | Nulstil opkaldspinkode<https://dialin.teams.microsoft.com/usp/pstnconferencing> ________________________________________________________________________________