Seminars
Seminars are held monthly and are open to all interested members of the scientific community.
Unless otherwise stated, these seminars will take place on Zoom Fridays once a month at 1:30-2:30 PM ET. Some webinars will be hybrid events with an in-person component at the 黑料社区 campus.
To join, you must be subscribed to our . You can also add the seminars to your calendar by subscribing to the Google Calendar below. Past recordings and information are available in the online archive and on the .
Upcoming Seminars
November 6, 2026 - WEBINAR
, Georgia Tech
Title: Trustworthy Frequentist Neural Simulation-Based Inference at Colliders
Abstract: We discuss two methods enabling reliable frequentist simulation-based inference. First, we introduce wifi ensembles, a framework using parametrized combinations of models to obtain asymptotic frequentist uncertainties on density ratios, with a particular focus on neural ratio estimation in the context of high-energy physics. When the density ratio of interest is a likelihood ratio conditioned on parameters, wifi ensembles can be used to perform simulation-based inference on those parameters while propagating the uncertainties on the density ratio. Then, we introduce Latent Categories, a method which uses a learned categorization to combine the reliability guarantees of histogram-based analyses with the improved sensitivity of simulation-based inference.
Bio: Sean Benevedes is a postdoctoral fellow at Georgia Tech developing ML-based statistical methods for collider physics and AI tools for model building. He completed his PhD at MIT in 2026 working with Jesse Thaler.
December 4, 2026 - WEBINAR
, University of Minnesota
Title: TBA
Abstract: TBA