Scientific committee
 Abstract submission
 Local information
 Research Unit 1735

In cooperation with


Keynote speakers are

  • S. Dasgupta (University of California, San Diego) giving a mini-course on "Learning with minimal supervision" (download slides 1st day, slides 2nd day, slides 3rd day) and
  • S. Mallat (École normale supérieure, Paris) giving a mini-course on "High Dimensional Classification with Deep Scattering Networks" (download slides).
Invited speakers are
  • J. Jin (Carnegie Mellon University), Fast Network Community Detection by SCORE
  • G. Kerkyacharian (Université Pierre et Marie Curie, Paris), From statistical estimation to the construction of wavelet, in a geometrical framework: The heat kernel point of view.
  • G. Lecué (CNRS and Université Paris-Est Marne-la-Vallée), Learning sub-Gaussian classes: upper and minimax bounds.
  • A. Munk (Georg August University Göttingen and Max Planck Institute for Biophysical Chemistry), Multiscale Change Point Inference
  • A. Nobel (University of North Carolina at Chapel Hill), Large Average Submatrices of a Gaussian Random Matrix: Landscapes and Local Optima.
  • A. Tsybakov (CREST-ENSAE, Paris), Empirical entropy, minimax regret and minimax risk

Contributed Talks:

  • A. Carpentier (University of Cambridge), Testing the regularity of a smooth signal
  • F. Enikeeva (INRIA Rhône-Alpes), High-dimensional change-point detection under sparsity assumptions
  • D. Hohmann (Philipps-Universität Marburg), Weighted angle Radon transform: A generalized SVD and minimax estimation in Gaussian white noise
  • E. Roquain (Université Pierre et Marie Curie), False discovery proportion control for Gaussian dependence in high dimension
  • N. Serdyukova (Georg-August-Universität Göttingen), Adaptive estimation under single-index constraint in a regression model
  • C. Strauch (Ruhr-Universität Bochum), Sharp adaptive drift estimation for ergodic diffusions in higher dimension
  • E. Tánczos (Eindhoven University of Technology), Adaptive Sensing for Estimation of Structured Sparse Sets

Contributed Posters:

  • A. Andresen (Weierstraß Institute Berlin, Research Unit 1735), Finite sample analysis of maximum likelihood estimators and convergence of the alternating procedure
  • G. Blanchard, F. Göbel, U. von Luxburg (University of Potsdam, University of Hamburg, Research Unit 1735), Data-adapted Tight Frames on Graphs
  • T. Bodnar, T. Dickhaus (Humboldt University of Berlin, Research Unit 1735), False Discovery Rate Control under Archimedean Copula
  • A. Bott (Technische Universität Darmstadt), Estimation of a distribution from data with small measurement errors
  • E. Burnaev, A. Zaytsev (Datadvance, IITP RAS), Properties of posterior distribution of parameters for Gaussian processes regression
  • H. Drees, N. Neumeyer, L. Selk (University of Hamburg, Research Unit 1735), Regression with irregular errors
  • M. Jirak (Humboldt University of Berlin, Research Unit 1735), Adaptive function estimation with one-sided errors
  • K. Jurczak, M. Reiß, A. Rohde (Ruhr University Bochum, Humboldt University of Berlin, Research Unit 1735), Structural inference for high-dimensional matrices
  • A. Kremer (University of Rostock), Discretely observed Markov jump processes with an absorbing state
  • S. Kurras, U. von Luxburg, G. Blanchard (University of Hamburg, University of Potsdam, Research Unit 1735), Merging Vertex Weights into Similarity Graphs
  • Z. Mohdeb (University Constantine 1, Algeria), Testing the equality of nonparametric regression curves based on Fourier coefficients
  • F. Müller (Technische Universität Darmstadt), Estimation of a regression function corresponding to latent variables
  • P. Nasiri (Payame Noor University, Iran), M-Step for Exponential Distribution with Presence of Outliers
  • M. Panov (Moscow Insitute of Physics and Technology), Non-asymptotic Bernstein-von Mises Theorem For Semiparametric Estimation
  • S. Schwaar (Technische Universität Kaiserslautern), Asymptotic Distribution of Change-Point Estimators in Nonlinear AR-Processes
  • M. Vollmer (University of Greifswald), Randomized Model Selection as a Model Building Strategy in Stepwise Logistic Regression

Mini-courses will take place on mornings, invited and contributed talks in the afternoon.

A visit of the "Neues Palais" or of the Park Sanssouci will be proposed to participants on the second conference day, this will be followed by conference dinner in the city of Potsdam.

A detailed schedule can be found here. The main conference room is building 11, Room 009 (ground floor).

Photo: Karla Fritze