Fraunhofer-Institut für Intelligente Analyse- und Informationssysteme IAIS

Filters

3
3
Reset filters

Settings
Now showing 1 - 3 of 3
  • Publication
    Explaining deep reinforcement learning decisions in complex multiagent settings: towards enabling automation in air traffic flow management
    ( 2023)
    Kravaris, Theocharis
    ;
    Lentzos, Konstantinos
    ;
    Santipantakis, Georgios
    ;
    Vouros, George A.
    ;
    Andrienko, Gennady
    ;
    Andrienko, Natalia
    ;
    Crook, Ian
    ;
    Garcia, Jose Manuel Cordero
    ;
    Martinez, Enrique Iglesias
    With the objective to enhance human performance and maximize engagement during the performance of tasks, we aim to advance automation for decision making in complex and large-scale multi-agent settings. Towards these goals, this paper presents a deep multi agent reinforcement learning method for resolving demand - capacity imbalances in real-world Air Traffic Management settings with thousands of agents. Agents comprising the system are able to jointly decide on the measures to be applied to resolve imbalances, while they provide explanations on their decisions: This information is rendered and explored via appropriate visual analytics tools. The paper presents how major challenges of scalability and complexity are addressed, and provides results from evaluation tests that show the abilities of models to provide high-quality solutions and high-fidelity explanations.
  • Publication
    SPARQL querying for validating the usage of automatically georeferenced social media data as human sensors for air quality
    ( 2022-07-11)
    Andreadis, Stelios
    ;
    Mavropoulos, Thanassis
    ;
    Pantelidis, Nick
    ;
    Vrochidis, Stefanos
    ;
    Elias, Mirette Magdy Michel
    ;
    Papadopoulos, Charis
    ;
    Gialampoukidis, Ilias
    ;
    Kompatsiaris, Ioannis
    The problem of air pollution is one of the countless topics discussed on social media on an everyday basis. This rich, crowdsourced information can be exploited to assess the air quality of urban areas, using humans as sensors. Nevertheless, the majority of social media data are falsely geotagged or completely lack geoinformation, which is an essential attribute, while the reliability of the air pollution events reported by online citizens has to be proven. The scope of this work is to present a framework that collects Twitter messages in German that refer to the atmosphere, automatically georeferences them, and finally validates them through semantic representation and SPARQL queries in order to associate them with real measurements of air quality sensors. The georeferencing models are evaluated against state-of-the-art works and the proposed framework is validated in a near-six-month scenario in Germany.
  • Publication
    Supporting Visual Exploration of Iterative Job Scheduling
    ( 2022-03-30)
    Andrienko, Gennady
    ;
    Andrienko, Natalia
    ;
    Garcia, Jose Manuel Cordero
    ;
    Hecker, Dirk
    ;
    Vouros, George A.
    We consider the general problem known as job shop scheduling, in which multiple jobs consist of sequential operations that need to be executed or served by appropriate machines having limited capacities. For example, train journeys (jobs) consist of moves and stops (operations) to be served by rail tracks and stations (machines). A schedule is an assignment of the job operations to machines and times where and when they will be executed. The developers of computational methods for job scheduling need tools enabling them to explore how their methods work. At a high level of generality, we define the system of pertinent exploration tasks and a combination of visualizations capable of supporting the tasks. We provide general descriptions of the purposes, contents, visual encoding, properties, and interactive facilities of the visualizations and illustrate them with images from an example implementation in air traffic management. We justify the design of the visualizations based on the tasks, principles of creating visualizations for pattern discovery, and scalability requirements. The outcomes of our research are sufficiently general to be of use in a variety of applications.