11-13 December 2017
Lawrence Berkeley National Laboratory
US/Pacific timezone

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

13 Dec 2017, 09:20
2-100 (Lawrence Berkeley National Laboratory)


Lawrence Berkeley National Laboratory


Bryan Ostdiek (University of Oregon)


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 are then trained on this modified data. Observed decreases in sensitivity diagnose the variable's discriminating power. Planing also allows the investigation of the linear versus non-linear nature of the boundaries between signal and background. I will demonstrate these features in both an easy to understand toy model and an idealized LHC resonance scenario.

Primary author

Bryan Ostdiek (University of Oregon)

Presentation Materials