Text only print
Full screen print
Today's seminar
Seminar information archive ～05/15｜Today's seminar 05/16 | Future seminars 05/17～
2025/05/16
Seminar on Probability and Statistics
13:30-14:30 Room #128 (Graduate School of Math. Sci. Bldg.)
Maud Delattre (INRAE)
Efficient precondition stochastic gradient descent for estimation in latent variables models (English)
https://u-tokyo-ac-jp.zoom.us/meeting/register/yixIylc3S8uJqOQ_Vqm_3Q
Maud Delattre (INRAE)
Efficient precondition stochastic gradient descent for estimation in latent variables models (English)
[ Abstract ]
Latent variable models are powerful tools for modeling complex phenomena involving in particular partially observed data, unobserved variables or underlying complex unknown structures. Inference is often difficult due to the latent structure of the model. To deal with parameter estimation in the presence of latent variables, well-known efficient methods exist, such as gradient-based and EM-type algorithms, but with practical and theoretical limitations. In this work, we propose as an alternative for parameter estimation an efficient preconditioned stochastic gradient algorithm.
Our method includes a preconditioning step based on a positive definite Fisher information matrix estimate. We prove convergence results for the proposed algorithm under mild assumptions for very general latent variable models. We illustrate through relevant simulations the performance of the proposed methodology in a nonlinear mixed-effects model.
[ Reference URL ]Latent variable models are powerful tools for modeling complex phenomena involving in particular partially observed data, unobserved variables or underlying complex unknown structures. Inference is often difficult due to the latent structure of the model. To deal with parameter estimation in the presence of latent variables, well-known efficient methods exist, such as gradient-based and EM-type algorithms, but with practical and theoretical limitations. In this work, we propose as an alternative for parameter estimation an efficient preconditioned stochastic gradient algorithm.
Our method includes a preconditioning step based on a positive definite Fisher information matrix estimate. We prove convergence results for the proposed algorithm under mild assumptions for very general latent variable models. We illustrate through relevant simulations the performance of the proposed methodology in a nonlinear mixed-effects model.
https://u-tokyo-ac-jp.zoom.us/meeting/register/yixIylc3S8uJqOQ_Vqm_3Q
Algebraic Geometry Seminar
13:30-15:00 Room #118 (Graduate School of Math. Sci. Bldg.)
Keita Goto (University of Tokyo)
Berkovich geometry and SYZ fibration
Keita Goto (University of Tokyo)
Berkovich geometry and SYZ fibration
[ Abstract ]
The SYZ fibration refers to a special Lagrangian torus fibration on a Calabi–Yau manifold and has been extensively studied in the context of mirror symmetry.
In particular, for a degenerating family of Calabi--Yau manifolds, a family of SYZ fibrations defined on each fiber, away from a subset of sufficiently small measure, plays a central role.
However, the existence of such fibrations remains an open problem, known as the metric SYZ conjecture.
To approach this problem, formal analytic techniques are particularly effective, and Berkovich geometry lies at their foundation.
In this talk, I will explain Yang Li’s "comparison property," a sufficient condition for the conjecture, and present some related results I have been involved in. Along the way, I will also introduce some foundational ideas in Berkovich geometry.
The SYZ fibration refers to a special Lagrangian torus fibration on a Calabi–Yau manifold and has been extensively studied in the context of mirror symmetry.
In particular, for a degenerating family of Calabi--Yau manifolds, a family of SYZ fibrations defined on each fiber, away from a subset of sufficiently small measure, plays a central role.
However, the existence of such fibrations remains an open problem, known as the metric SYZ conjecture.
To approach this problem, formal analytic techniques are particularly effective, and Berkovich geometry lies at their foundation.
In this talk, I will explain Yang Li’s "comparison property," a sufficient condition for the conjecture, and present some related results I have been involved in. Along the way, I will also introduce some foundational ideas in Berkovich geometry.
