Bachelor Thesis Open Access
Oßwald, Paul
{ "conceptrecid": "22216", "created": "2024-02-13T10:04:03.859344+00:00", "files": [ { "bucket": "834470c9-899d-4a58-abbe-3242be944415", "checksum": "md5:bd063daba99ddf1d6075ab2e77414d3b", "key": "Bachelorthesis_Paul_Osswald_GNNExplainer.pdf", "links": { "self": "https://publish.etp.kit.edu/api/files/834470c9-899d-4a58-abbe-3242be944415/Bachelorthesis_Paul_Osswald_GNNExplainer.pdf" }, "size": 2338128, "type": "pdf" } ], "id": 22217, "links": { "bucket": "https://publish.etp.kit.edu/api/files/834470c9-899d-4a58-abbe-3242be944415", "html": "https://publish.etp.kit.edu/record/22217", "latest": "https://publish.etp.kit.edu/api/records/22217", "latest_html": "https://publish.etp.kit.edu/record/22217" }, "metadata": { "access_right": "open", "access_right_category": "success", "communities": [ { "id": "cms" }, { "id": "etp" } ], "contributors": [], "creators": [ { "affiliation": "KIT/ETP", "name": "O\u00dfwald, Paul" } ], "description": "<p>In the last years, efforts have been made to use various machine learning models in the multiclass classification of the tt+X events. Especially Graph Neural Networks (GNN) architectures have received attention, due to the data structure lending itself to the representation as a multi-relational graph. Naturally, when using a GNN to classify an event, it is of interest why it makes a particular prediction. GNNs however, due to the complex structure of graph data and black-box nature, are hard to interpret.</p>\n\n<p>This thesis focuses on examining a GNN trained for this classification task using the GNNExplainer method to better understand the underlying decisions taken by the GNN when using different tt+X events as input.</p>", "keywords": [ "Graph Neural Networks", "GNNExplainer", "CMS", "ttX", "Machine learning" ], "language": "eng", "publication_date": "2023-11-08", "relations": { "version": [ { "count": 1, "index": 0, "is_last": true, "last_child": { "pid_type": "recid", "pid_value": "22217" }, "parent": { "pid_type": "recid", "pid_value": "22216" } } ] }, "resource_type": { "subtype": "bachelor-thesis", "title": "Bachelor Thesis", "type": "thesis" }, "thesis": { "report_number": "ETP-BACHELOR-KA/2023-13", "supervisors": [ { "affiliation": "KIT/ETP", "name": "Husemann, Prof. Dr. Ulrich" }, { "affiliation": "KIT/ETP", "name": "Wassmer, Dr. Michael" }, { "affiliation": "KIT/ETP", "name": "Pfeffer, Emanuel" } ] }, "title": "Generating Explanations for Graph Neural Networks for tt+X events at the CMS Experiment" }, "owners": [ 79 ], "revision": 3, "stats": {}, "updated": "2024-02-13T10:04:06.015866+00:00" }