b-jet GNN Tagging
Development of GNN-based methods for b-jet tagging in heavy-ion environments.
This research focuses on improving the identification of bottom-quark jets using Graph Neural Networks (GNNs).
Research Highlights
- Optimized GNN architecture for high-multiplicity backgrounds.
- Performance evaluation in oxygen-oxygen and lead-lead collision simulations.
- Comparison with traditional CNN and BDT-based tagging methods.