UnRAVeL Herbst Workshop

Donnerstag, 16. & Freitag, 17. November 2023

 

Der UnRAVeL Herbstworkshops steht unter dem Motto "Unsere eigenen Potenziale erkennen". Wir besuchen die Einrichtungen unserer Mitglieder und laden ehemalige UnRAVeLs zum Erfahrungsaustausch ein.

 

Programm

Thursday 16.11.2023
9:00 Start in Raum 9222, Begrüßung und Vorstellung der Aktivitaten und Gremien
9:30 Introtalks der PhDs
10:30 Pause
10:45 Introtalks PhD 2
12:30 Mittagessen
13:30 Introtalks der PIs
14:00

Talks der PhDs

Tamme Emunds: To build or not to build: Estimating the effect of overpass structures for railway junctions

Many infrastructure managers have the goal to increase the capacity of their railway infrastructure due to an increasing demand. While methods for performance calculations of railway line infrastructure are already well established, the determination of railway junction capacity remains a challenge. This work utilizes the concept of queueing theory to develop a method for the capacity calculation of railway junctions, solely depending on their infrastructure layout along with arrival and service rates. The implementation of the introduced approach is based on advanced model-checking techniques. It can be used to decide which infrastructure layout to build, i.e. whether an overpass for the analyzed railway junction is needed. The developed method hence addresses the need for fast and reliable timetable-independent junction evaluation in the long-term railway capacity calculation landscape.

----------------------------

Tobias Winkler: Programmatic Strategy Synthesis: Resolving Nondeterminism in Probabilistic Programs

Markov decision processes (MDPs) are a standard model for sequential decision making and planning in probabilistic environments. MDPs are applied in fields such as robotics, economics, game theory, reinforcement learning, and others. The basic MDP problem is to find a decision rule, also called strategy or policy, that performs well according to some criteria. Numerous methods for constructing good or even optimal strategies for a given MDP exist. In this talk, we are concerned with finding strategies for infinite families of structurally similar MDPs. For example, in a gambling game we may not know our initial budget upfront, and we would like to have a parameterized strategy for all possible budgets. We encode such parameterized MDPs as simple imperative programs enhanced with coin flips and nondeterminstic branching — the latter is used to model the decisions. Our approach, which draws on principles from deductive program verification, most prominently pre- and postconditions as well as loop invariants, resolves these decisions with provable performance guarantees.
This talk is based on joint work with Kevin Batz, Tom Biskup, and Joost-Pieter Katoen.

14:45 Kaffeepause
15.00

Event: Eisenbahnlehr- und Versuchsanlage (@ VIA)

17:00 Ende
20:00 opt: Get together: Kegeln im Zuhause
Friday 17.11.2023
8:30 Event: DSME, Treffpunkt: Foyer des TZA, Dennewartstr 25-27
10:00

Pause und Spaziergang zur Digi Church

10:15 Start in Digital Church, Jülicher Str. 72a, 52070 Aachen

PhD Talk
Vincent Grande: Topological Point Cloud Clustering: Taking spectral clustering to the next dimension

We present Topological Point Cloud Clustering (TPCC), a new method to cluster points in an arbitrary point cloud based on their contribution to global topological features. TPCC synthesizes desirable features from spectral clustering and topological data analysis and is based on considering the spectral properties of a simplicial complex associated to the considered point cloud. As it is based on considering sparse eigenvector computations, TPCC is similarly easy to interpret and implement as spectral clustering. However, by focusing not just on a single matrix associated to a graph created from the point cloud data, but on a whole set of Hodge-Laplacians associated to an appropriately constructed simplicial complex, we can leverage a far richer set of topological features to characterize the data points within the point cloud and benefit from the relative robustness of topological techniques against noise.

Additionally selected talks of our UnRAVeL Alumni.

12:30 Mittagessen
13:30

Austausch mit UnRAVeL Alumni: Dr. Marcel Hark, Prof. Dr. Benjamin Kaminski, Dr. Martin Ritzert, Stephan Zieger, Dr. Rebecca Haehn, Dr. Björn Tauer

15:30 Split feedback on UnRAVeL
16:00 Feedbackrunde
16:30 Zusammenfassung
17:00 Ende