Abstract
Applications of machine learning tools to problems of physical interest are often criticized for producing sensitivity at the expense of transparency. To address this concern, we explore a data planing 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. We demonstrate the efficacy of this approach using a toy example, followed by an application to an idealized heavy resonance scenario at the Large Hadron Collider. By unpacking the information being utilized by these algorithms, this method puts in context what it means for a machine to learn.
Highlights
A common argument against using machine learning for physical applications is that they function as a black box: send in some data and out comes a number
While this kind of nonparametric estimation can be extremely useful, a physicist often wants to understand what aspect of the input data yields the discriminating power, in order to learn/ confirm the underlying physics or to account for their systematics
III we show that these features can be realized in a more realistic particle physics setting
Summary
A common argument against using machine learning for physical applications is that they function as a black box: send in some data and out comes a number. We use the “uniform phase space” scheme to flatten discriminating variables, which was introduced in [6] to quantify the information learned by deep neural networks. It is possible to plane away all the underlying discriminating characteristics of this toy by utilizing combinations of linear and nonlinear variables This highlights another salient attribute of data planing: by comparing the performance of linear and deep neural networks, one can infer to what extent the encoded information is a linear versus nonlinear function of the inputs. IV concludes this paper with a discussion of future investigations
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
