## October 2019

### Eindhoven Stochastics Seminar

Joost Jorritsma (Eindhoven University of Technology) Typical weighted distance in preferential attachment models - Interpolating small and mini worlds Information diffusion in networks can be modeled by first passage percolation, which can also be seen as an SI (susceptible - infected) model. In this talk I will discuss the model for two (non-spatial) preferential attachment models where the degrees follow a power law with parameter in (2, 3): once a preferential attachment graph on N vertices is created, we equip…

Find out more »## November 2019

### Eindhoven Stochastics Seminar

Tim Hulshof (TU Eindhoven) A scaling limit for random walk in a very restrictive random environment Random walk on a lattice has Brownian motion as its scaling limit. Random walk on a slightly damaged lattice also has a Brownian scaling limit. But when the damage is so severe that the lattice (almost) falls apart into only finite pieces, the random walk suddenly moves orders of magnitude more slowly, and the scaling limit (if it exists) must have dramatically different properties.…

Find out more »### Eindhoven Stochastics Seminar

(please note: different room!!) Johannes Schmidt-Hieber (UTwente) Towards a statistical foundation of deep learning Recently a lot of progress has been made in the theoretical understanding of deep learning. One of the very promising directions is the statistical approach, which interprets deep learning as a statistical method and builds on existing techniques in mathematical statistics to derive theoretical error bounds. The talk surveys this field and describes future challenges.

Find out more »## December 2019

### Eindhoven Stochastics Seminar

Kay Bogerd (TU Eindhoven) Detecting small communities in inhomogeneous random graphs Many networks are organized in communities, with certain groups of vertices being more densely connected than others. When these communities are large enough they can easily be detected, whereas small communities can be more difficult, especially when the network is also inhomogeneous. In this talk, we consider the problem of detecting whether an inhomogeneous random graph contains a small community. We show when a scan test can asymptotically detect…

Find out more »### Workshop "Heavy Tails"

Summary The goal of the workshop is to bring together researchers from probability, statistics, and various application areas such as computer science, operations research, physics, engineering and finance and learn from each other on the latest developments in theory , statistical and simulation algorithms, and applications. Sponsors Organizers Remco van der Hofstad TU Eindhoven Adam Wierman Caltech Bert Zwart CWI / TU Eindhoven Speakers August Balkema University of Amsterdam Bojan Basrak University of Zagreb Mihail Bazhba CWI…

Find out more »### Eindhoven Stochastics Seminar

Gianmarco Bet (University of Florence) Weighted Dyck paths for non stationary queues We consider a model for a transitory queue in which only a fixed number $N$ of customers can join. Each customer joins the queue independently of the other ones at a random time, which we assume to be exponentially distributed. Assuming further that the service times also follow an exponential distribution, this system can be described as a two-dimensional Markov process on a triangular subset of $\mathbb Z^2$. The…

Find out more »## January 2020

### Eindhoven Stochastics Seminar

Seva Shneer (Heriot-Watt University) Utility-maximising properties of some service-rates allocations in wireless communication networks We consider wireless communication networks where service rates of nodes depend on their signal to interference plus noise ratios. We demonstrate that these rates are utility-maximising in a certain sense. Using this observation, we prove stability of single-hop networks under natural assumptions. We also obtain bounds on some moments of the stationary workload. In order to obtain these results, we consider a framework of utility-optimising rate allocations which is…

Find out more »## February 2020

### Eindhoven Stochastics Seminar

Jim Portegies (TU Eindhoven) Tropical probability and information optimization Together with Slava Matveev, we've been developing a theory of tropical probability, which is a metric-geometric perspective on information theory that can be helpful in the study of information-optimization problems. In this talk I will give a coarse overview of what we have developed so far, explain what we mean by information-optimization problems and indicate an application to such a problem.

Find out more »### Eindhoven Stochastics Seminar

Rob van der Mei (CWI & VU Amsterdam) Saving Lives by Stochastic Optimization In serious, life-threatening situations, the timely presence of emergency services (ambulance, firefighters) can make the difference between survival or death. In this talk, I will discuss a number of models and optimization results and show that they can really make the difference.

Find out more »## March 2020

### Eindhoven Stochastics Seminar

Wouter de Vries (KPN) Artificial intelligent processes for people About 75 years after the first prudent attempts to capture artificial intelligence in conceptual models, the field is now an exuberant and flourishing domain. Many ingenious applications find their way to consumers and businesses, as well as the telecommunications sector. Through some appealing examples, such as image recognition and speech recognition, this lecture gives an impression of what artificial intelligence is about. In addition, some insights are given from the founding…

Find out more »