Research project
Machine Learning for Top Quark Effective Theory
Learning statistical structure in particle physics.
How can learned likelihood information help constrain new-physics couplings in top quark effective theory?
Approach
Integrate neural networks with the Standard Model Effective Field Theory framework and analyze learned likelihood scores through Fisher information.
My contribution
- Integrated neural networks with the SMEFT framework.
- Used learned likelihood scores and Fisher information to tighten coupling constraints.
Results
ML × SMEFTLearned scores and Fisher information
Learned likelihood scores and Fisher information yield tighter constraints on new-physics couplings in the SMEFT framework.
Research conducted at Fudan University with Prof. Jiayin Gu.