New framework analyzes neural network training for PDF fitting

Researchers have developed a theoretical framework using the Neural Tangent Kernel (NTK) to analyze the training dynamics of neural networks used in Parton Distribution Function (PDF) fitting. This approach offers an analytical description of neural network evolution during training, clarifying the impact of architecture and experimental data. It also provides a quantitative method for understanding how uncertainties propagate from data to the fitted functions, serving as a diagnostic tool for fitting methodologies. AI

IMPACT Provides a new analytical tool for understanding and validating machine learning models used in scientific research, potentially improving the robustness of data fitting and uncertainty quantification.

RANK_REASON The cluster contains an academic paper detailing a new theoretical framework for analyzing neural network training dynamics in the context of particle physics. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.LG →

• Amedeo Chiefa
• LHC
• Machine Learning
• Neural Networks
• Neural Tangent Kernel
• NNPDF
• Parton Distribution Functions

• paper
• infra

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →