Colloquia are typically hosted inHalligan Hall, room 102, on Thursdays from 2:50 - 4:00 pm. Refreshments are at 2:50, and talk starts at 3:00 pm. Please check details below to confirm times and locations.
To subscribe to CS colloquium announcements, or for any comments or questions about CS colloquia, please email colloquia@cs.tufts.edu.
October 11 at 3:00 PM: Generating New Knowledge in Art History and Material Culture with Technology: Case Studies from the Field and the Classroom
Title: Generating New Knowledge in Art History and Material Culture with Technology: Case Studies from the Field and the Classroom Speaker: Caroline Bruzelius, Duke University Date: Thursday, October 11, 2018 Time: 3:00 PM Location: Halligan 102 Host: Susan Landau
October 12 at 10:00 AM: Doctoral Thesis Defense: Algorithms and Theory for Variational Inference in Two-level Non-conjugate Models
Work over the last two decades has developed powerful models in probabilistic machine leaning with a variety of applications: recommender systems, statistical analyses of text corpora, automated astronomy, and more. Exact inference is rarely possible in complex models and much research has been devoted to developing successful algorithms for approximate inference. Variational approximations are one of the main paradigms for approximate inference but optimization of the variational approximation is challenging, and the theoretical properties of such approximations are not fully understood.
In this talk, we focus on variational inference in a rich, two-level latent variable family that includes as members sparse Gaussian processes, correlated topic models, probabilistic matrix factorization, and mixed effects models. The first part of the talk describes recently developed algorithmic tools for approximate inference in the presence of non-conjugacy within these types of models. The ideas here extend the popular stochastic variational inference paradigm to a far larger class of models. The second part of the talk focuses on the theoretical properties of variational inference. Specifically, we discuss several variants of the variational approximation and show that some of these have strong guarantees against the best non-Bayesian procedures.
October 18 at 3:00 PM: Developing Inclusive Digital Environments: Challenges and Opportunities for Computer Science
Title: Developing Inclusive Digital Environments: Challenges and Opportunities for Computer Science Speaker: Judy Brewer, MIT and W3C Date: Thursday, October 18, 2018 Time: 3:00 PM Location: Halligan 102 Host: Susan Landau
October 25: Data-driven Approaches for Uncovering Functional Variations in Protein Interactions
Title: Data-driven Approaches for Uncovering Functional Variations in Protein Interactions Speaker: Shilpa Nadimpalli Kobren, Harvard University Date: Thursday, October 25, 2018 Time: 3:00 PM Location: Halligan 102 Host: Lenore Cowen
November 1 at 3:00 PM: Colloquium Speaker Sarah Cannon, Georgia Tech