A Google Summer of Code 2022 Project Repository.
The End-to-End Deep Learning (E2E) project in the CMS experiment focuses on the development of these reconstruction and identification tasks with innovative deep learning approaches. This project will focus on the development of end-to-end graph neural networks for particle (tau) identification and CMSSW inference engine for use in reconstruction algorithms in offline and high-level trigger systems of the CMS experiment.
Item | Link |
---|---|
Organization | Machine Learning for Science (ML4SCI) |
Contributor | Xin Yi |
Project Details | Organization Project Page Accepted Proposal GSOC Project Page |
Notebook Version Name | Graph Representation | Test AUC |
---|---|---|
Graph Attention | K Nearest neighbors(k=25) | 0.8279 |
Graph Convolution | K Nearest neighbors(k=15) | 0.8593 |
Graph SAGE | K Nearest neighbors(k=15) | 0.8624 |
Dynamic Edge Convolution | Dynamic K Nearest neighbors(k=20) | 0.8627 |
Classical CNN | Image Data | Overfitting |