Bachelor Thesis Open Access

Generating Explanations for Graph Neural Networks for tt+X events at the CMS Experiment

Oßwald, Paul

Citation Style Language JSON Export

{
  "abstract": "<p>In the last years, efforts have been made to use various machine learning models in the multiclass classification of the tt+X&nbsp;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&nbsp;events as input.</p>", 
  "author": [
    {
      "family": "O\u00dfwald, Paul"
    }
  ], 
  "id": "22217", 
  "issued": {
    "date-parts": [
      [
        2023, 
        11, 
        8
      ]
    ]
  }, 
  "language": "eng", 
  "title": "Generating Explanations for Graph Neural Networks for tt+X events at the CMS Experiment", 
  "type": "thesis"
}
Publication date:
November 8, 2023
Keyword(s):
Graph Neural Networks GNNExplainer CMS ttX Machine learning
Awarding University:
Karlsruhe Institute of Technology (KIT)
Report Number:
ETP-BACHELOR-KA/2023-13
Communities:

Cite as

Export