Archiv: Stochastik Rhein-Main

08.02.2018 Oberseminar Darmstadt

Yulia Petrova (University St. Petersburg)

Exact L_2-small ball probabilities for finite-dimensional perturbations of Gaussian processes: spectral method

28.05.2019 Oberseminar Mainz

Vlada Limic (Université de Strasbourg)

20 years of the Multiplicative coalescent

20.06.2024 Oberseminar Darmstadt

Tejas Iyer (WIAS Berlin)

Persistent hubs in generalised preferential attachment trees

11.05.2017 Oberseminar Darmstadt

Stéphane Le Roux (Université libre de Bruxelles)

Stable Nash Equilibria

28.01.2020 Oberseminar Mainz

Stephan Gufler (Technion Haifa)

28.01.2020 Stephan Gufler (Technion Haifa)

28.05.2024 Oberseminar Mainz

Stein Andreas Bethuelsen (University of Bergen)

Mixing for Poisson representable processes and the contact process

In this talk I will present some new insights on so-called Poisson representable processes, a general class of {0,1}-valued processes recently introduced by Forsström, Gantert and Steif. Particularly, I will discuss a new characteristic of these in terms of certain mixing properties. As an application thereof, I will argue that the upper invariant measure of the contact process on Z^d is not Poisson representable, thereby answering a question raised in the above mentioned work. This relies on the upper invariant measure satisfying certain directional mixing properties, but not their spatial equivalent. Moreover, the general approach extends to other processes having similar properties, such as the plus phase of the Ising model on Z^2 in the phase transition regime. weiterlesen

21.06.2024 Rhein-Main-Kolloquium

Stein Andreas Bethuelsen (University of Bergen)

RMK Mainz

Assign to each lattice point of Z^d a Poissonian number of particles and let each of them evolve independently as discrete-time simple random walks. On top of this dynamically evolving environment we consider an additional random walk whose jump transition depends on whether there are particles present at its location or not. This is the so-called random walk on random walks model. Previous studies have concluded that this model on Z^d, d\geq1, has a diffusive scaling when the density of particles is sufficiently low or sufficiently high. We will argue that this holds for all densities for the model on Z^d with d\geq 5. Our proof of this rely on a novel domination result for the dynamic environment that, when combined with coupling arguments and standard random walk estimates, yield uniform mixing bounds for the so-called local environment process. Based on joint work, partly in progress, with Florian Völlering (University of Leipzig) weiterlesen

31.01.2019 Oberseminar Darmstadt

Steffen Dereich (Universität Münster)

Quasi-processes for branching Markov chains

11.05.2023 Oberseminar Darmstadt

Stefan Grosskinsky (Universität Augsburg)

Size-biased diffusion limits for the inclusion process

21.12.2017 Oberseminar Darmstadt

Sebastian Schwinn (TU Darmstadt)

Improving the performance of polling models using forced idle times

26.01.2017 Oberseminar Darmstadt

Sebastian Riedel (Universität Berlin)

TBA

16.04.2019 Oberseminar Mainz

Sandra Kliem (Goethe-Universität Frankfurt)

Representations for measure-valued branching processes revisited

04.06.2024 Oberseminar Mainz

Samuel Modee (University of Bergen)

Some rigorous results on the zombie infection model

The so-called zombie infection model is a one-parameter family of interacting particle systems introduced in the physics literature in the 2010’s. This model can be seen as a natural hybrid between a SIR infection model and the bias-voter model. After a brief survey on the basics of this model, we present in this talk some rigorous results, focusing on monotonicity properties and bounds on the probability of a zombie invasion. This is based on joint work, in progress, with Erik Broman and Stein Andreas Bethuelsen. weiterlesen

04.07.2017 Oberseminar Mainz

Roger Tribe (University of Warwick)

Coalescing Random Walks and related models

25.04.2024 Oberseminar Darmstadt

Ralph Neininger (Goethe-Universität Frankfurt)

Recursive distributional equations in applications

23.06.2022 Oberseminar Darmstadt

Quirin Vogel (NYU Shanghai)

The limit of the Feynman representation for the Bose gas

16.05.2017 Stochastik-Kolloquium Frankfurt

Prof. Dr. Sylvie Roelly (Universität Potsdam)

Die Renaissance des Werkes des Mathematikers Vincent Doeblin

07.02.2023 Oberseminar Mainz

Prof. Dr. Steffen Dereich (Universität Münster)

Oberseminar Stochastik Mainz - 07.05.2024 - Prof. Dr. Steffen Dereich (WWU Münster)

Titel: New error bounds for the Adam optimisation algorithm Abstract: In this talk, I will give an introduction concerning the training of artificial neural networks based on stochastic gradient descent algorithms. Whereas there are various approaches for carrying out an error analysis for classical stochastic gradient descent, significantly less is known for more elaborate algorithms such as the Adam algorithm. Recently, we established a new approach which allows us to provide error estimates and to reduce this gap. The talk is based on joint work with Arnulf Jentzen. weiterlesen

07.05.2024 Oberseminar Mainz

Prof. Dr. Steffen Dereich (Universität Münster)

New error bounds for the Adam optimisation algorithm

In this talk, I will give an introduction concerning the training of artifcial neural networks based on stochastic gradient descent algorithms. Whereas there are various approaches for carrying out an error analysis for classical stochastic gradient descent, significantly less is known for more elaborate algorithms such as the Adam algorithm. Recently, we established a new approach which allows us to provide error estimates and to reduce this gap. The talk is based on joint work with Arnulf Jentzen. weiterlesen

12.06.2018 Stochastik-Kolloquium Frankfurt

Prof. Dr. Sebastian Mentemeier

Limit theorems for recursive cell-splitting schemes

Veranstaltungsarchiv

Events

Exact L_2-small ball probabilities for finite-dimensional perturbations of Gaussian processes: spectral method

20 years of the Multiplicative coalescent

Persistent hubs in generalised preferential attachment trees

Stable Nash Equilibria

28.01.2020 Stephan Gufler (Technion Haifa)

Mixing for Poisson representable processes and the contact process

RMK Mainz

Quasi-processes for branching Markov chains

Size-biased diffusion limits for the inclusion process

Improving the performance of polling models using forced idle times

TBA

Representations for measure-valued branching processes revisited

Some rigorous results on the zombie infection model

Coalescing Random Walks and related models

Recursive distributional equations in applications

The limit of the Feynman representation for the Bose gas

Die Renaissance des Werkes des Mathematikers Vincent Doeblin

Oberseminar Stochastik Mainz - 07.05.2024 - Prof. Dr. Steffen Dereich (WWU Münster)

New error bounds for the Adam optimisation algorithm

Limit theorems for recursive cell-splitting schemes

Überblick

Kontakt