We study fundamental mathematical problems originating form the field of Network Optimization.
There are plenty of networks (e.g. electrical, communication, transportation, computer networks), whose efficiency is crucial. Typically, the geometry of the underlying problem is modeled as a graph G=(V,E) and the problem is to install elements of the constructed network in vertices and edges of G. Constructing an element incurs a certain cost which may depend on the location in G. We aim at obtaining a minimal cost network having all the required properties.
Such interesting network problems are mostly NP-hard, so it is unlikely that efficient algorithms for finding optimal solutions for these problems exist. We concentrate on efficiently finding close to optimal solutions by means of approximation algorithms.
An important field of network optimization problems is related to Network Design problems, where we are given a set of nodes of the graph (called terminals) that want to communicate with each other, and we need to buy some edges of the graph to connect all the terminals. Famous examples could be Steiner Tree problem, or Virtual Private Network design problems. Still, in case of economies of scale in buying capacities, the total installation cost could be modeled as a concave function: classical examples are buy-at-bulk problems or rent-or-buy problems. We concentrate on developing constant approximation algorithms for such problems and settle some important compexity issues.
Another field is related to Location problems: a relevant example is the Facility Location Problem, where we need to install facilities in vertices of the graph in order to service a given set of clients. Opening a facility incurs a location specific cost, servicing a client with a facility contributes a cost proportional to the client-facility distance.
We studied the basic variant of the problem called metric Uncapacitated Facility Location, and the more general Connected Facility Location problem, where we also need to connect all the facilities, which incurs cost proportional to the total length of the edges used to connect them. We give constant factor approximation algorithms for both the variants.
A Politics of Intellectual Property: The French Revolution and the Creation of a Patent SystemTechnology & Culture. 2020-07-01.
Understanding multimodality through rhythm of life. An empirical evidence from the Swiss case study.Handbook of urban mobilities; Routledge, 2020-01-01.
Exact synthesis of LTL properties from traces2019-12-31. Forum on specification & Design Languages – FDL 2019, Southampton, UK, September 2-4, 2019.
Current Developments in Biotechnology and Bioengineering: Resource Recovery from WastesElsevier, 2019-11-30.
AMS-100: The next generation magnetic spectrometer in space – An international science platform for physics and astrophysics at Lagrange point 2Nuclear Instruments & Methods In Physics Research Section A-Accelerators Spectrometers Detectors And Associated Equipment. 2019-11-11. Vol. 944, p. 162561. DOI : 10.1016/j.nima.2019.162561.
Refining Biomass Residues for Sustainable Energy and Bioproducts: Technology, Advances, Life Cycle Assessment and EconomicsElsevier Science Publishing Co Inc, 2019-11-01.
SynDeMo: Synergistic Deep Feature Alignment for Joint Learning of Depth and Ego-Motion2019-10-27. 2019 International Conference on Computer Vision (ICCV 2019), Seoul, South Korea, 27-10, 2019.
Development and validation of a double focalizing magnetic spectrometer for beta spectrum measurementsNuclear Instruments & Methods In Physics Research Section A-Accelerators Spectrometers Detectors And Associated Equipment. 2019-10-21. Vol. 942, p. 162384. DOI : 10.1016/j.nima.2019.162384.
Importance of the advection scheme for the simulation of water isotopes over Antarctica by atmospheric general circulation models: A case study for present-day and Last Glacial Maximum with LMDZ-isoEarth and Planetary Science Letters. 2019-10-15. Vol. 524, p. 115731. DOI : 10.1016/j.epsl.2019.115731.
Exploration of logics underlying modal choice of the working population from the Greater Geneva2019-10-14. Swiss Mobility Conference, Lausanne, Vaud, Switzerland, October 14, 2019.
Advanced Timber Construction, using digital fabrication and robotic assembliesGraduates Seminar Talks, University of Southern California, USA, 14.10.19.
L’évolution des modes de vie et les ateliers d’autoréparation de véloAssemblée Générale du P’tit vélo dans la tête, Grenoble,
Impact of advanced electricity tariff structures on the optimal design, operation and profitability of a grid-connected PV system with energy storage8th DACH+ Conference on Energy Informatics, Salzburg, September 26-27, 2019.