The NNPDF collaboration determines the structure of the proton using contemporary methods of artificial intelligence. A precise knowledge of the so-called Parton Distribution Functions (PDFs) of the proton, which describe their structure in terms of their quark and gluon constituents, is a crucial ingredient of the physics program of the Large Hadron Collider of CERN. It has played an important role in the discovery of the Higgs boson. Its incomplete knowledge is one of the main limitations in searches of new physics.
PDFs cannot be computed from first principles: they have to be extracted from the data, through a careful comparison of theoretical predictions and experimental results. NNPDF determines PDFs using as an unbiased modeling tool Neural Networks, trained using Genetic Algorithms, and used to construct a Monte Carlo representation of PDFs and their uncertainties: a probability distribution in a space of functions.
This site provides information on NNPDF for the general public, for physicists, and for PDF users. Among others, a description of our main research tools, user manuals and documentation, talks and publications, including theses, and links to analysis tools. All NNPDF PDF sets are publicly available from LHAPDF.