Upcoming Events

Many events are currently organized online. Information on how to access these events can be found by clicking “more” below the respective entry.


go to archive

Monday, 05.12.2022, 10:30 (WIAS-406)
Seminar in Machine Learning and Data-Driven Optimization
Yucen Luo, Max Planck Institute for Intelligent Systems:
Spectral representation learning for conditional moment models (hybrid talk)
more ... Location
Weierstraß-Institut, Mohrenstr. 39, 10117 Berlin, 4. Etage, Weierstraß-Hörsaal (Raum: 406)

Abstract
Many problems in causal inference and economics can be formulated in the framework of conditional moment models, which characterize the target function through a collection of conditional moment restrictions. For nonparametric conditional moment models, efficient estimation has always relied on preimposed conditions on various measures of ill-posedness of the hypothesis space, which are hard to validate when flexible models are used. In this work, we address this issue by proposing a procedure that automatically learns representations with controlled measures of ill-posedness. Our method approximates a linear representation defined by the spectral decomposition of a conditional expectation operator, which can be used for kernelized estimators and is known to facilitate minimax optimal estimation in certain settings. We show this representation can be efficiently estimated from data, and establish $L_2$ consistency for the resulting estimator. We evaluate the proposed method on proximal causal inference tasks, exhibiting promising performance on high-dimensional, semi-synthetic data.

Further Informations
The talk is also given via Zoom: https://us02web.zoom.us/j/8994552597

Host
WIAS Berlin
Tuesday, 06.12.2022, 15:00 (WIAS-405-406)
Seminar Modern Methods in Applied Stochastics and Nonparametric Statistics
Dr. Alexandra Suvorikova, WIAS Berlin:
Anomaly detection in biometric authentication (hybrid talk)
more ... Location
Weierstraß-Institut, Mohrenstr. 39, 10117 Berlin, 4. Etage, Raum: 405/406

Abstract
In this work we suggest a novel framework to detect user behaviour anomalies based on transfer learning approach combined with optimal transport techniques.

Further Informations
Dieser Vortrag findet bei Zoom statt: https://zoom.us/j/492088715

Host
WIAS Berlin
Wednesday, 07.12.2022, 11:30 (WIAS-405-406)
Seminar Interacting Random Systems
Daniel Heydecker, MPI Leipzig:
From LDPs of the Kac Process to the H-Theorem
more ... Location
Weierstraß-Institut, Mohrenstr. 39, 10117 Berlin, 4. Etage, Raum: 405/406

Abstract
It is well-recognised that the large deviations of particle systems are connected to the appearance of a gradient flow formulation through a reformulation of the dynamical large deviations cost. In this talk, I will discuss a simple proof of Boltzmann's H-Theorem in the spatially homogeneous Boltzmann equation, which yields the equality between entropy and integrated entropy dissipation under quite general conditions. Time allowing, I will also discuss recent work applying the same principles to the porous medium equation.

Further Informations
Seminar Interacting Random Systems

Host
WIAS Berlin