Speaker
Description
The large samples of Z0 bosons expected at the potential Future Circular Collider (FCC-ee) will provide an unprecedented environment for precision flavour physics. In this context, the ability to reconstruct heavy-hadron decay chains with high efficiency and purity is essential to fully exploit the physics potential of the experiment, enabling improved measurements of rare decays, CP violation, and heavy-flavour fragmentation properties.
A promising strategy is the use of a Deep-learning based Full Event Interpretation (DFEI), designed to perform the simultaneous identification, isolation, and hierarchical reconstruction of all heavy-hadron decay chains in each event. Such an approach goes beyond traditional reconstruction pipelines by directly targeting global event interpretation, combining tracking, vertexing, and decay topology information in a unified framework.
In this work, we present a proposal for a graph neural network (GNN) architecture tailored to FCC-ee conditions using IDEA detector simulation, aimed at achieving full heavy-hadron reconstruction through an end-to-end learning strategy. The proposed method exploits the natural graph structure of particle collisions, enabling the reconstruction of complex multi-body decay chains.
| I read the instructions above | Yes |
|---|