Machine Learning for Jet Physics

Session

Learning from data

13 Dec 2017, 09:00

2-100 (Lawrence Berkeley National Laboratory)

25. Introduction and Overview (15'+5')

Marat Freytsis (University of Oregon)

13/12/2017, 09:00

20. "Planing" to expose what the machine is learning (15'+5')

Bryan Ostdiek (University of Oregon)

13/12/2017, 09:20

Applications of machine learning tools to problems of physical interest are often criticized for producing sensitivity at the expense of transparency. In this talk, I explore a procedure for identifying combinations of variables -- aided by physical intuition -- that can discriminate signal from background. Weights are introduced to smooth away the features in a given variable(s). New networks...

22. Weak Supervision in High Dimensions (15'+5')

Eric Metodiev (MIT)

13/12/2017, 09:40

21. Building an anti-QCD tagger (15'+5')

Jack Collins (University of Maryland)

13/12/2017, 10:00

23. Adversarial Approaches (15'+5')

Kyle Cranmer (NYU)

13/12/2017, 10:20

We use an adversarial neural network to train a jet classifier that remains largely uncorrelated with the jet mass --- a nuisance parameter that is highly correlated with the observed features. This adversarial training strategy balances the dual objectives of classification accuracy and decorrelation, reducing the deleterious effect of systematic uncertainties in the background modeling. The...