Search for: All records

A bstract A framework is presented to extract and understand decision-making information from a deep neural network (DNN) classifier of jet substructure tagging techniques. The general method studied is to provide expert variables that augment inputs (“eXpert AUGmented” variables, or XAUG variables), then apply layerwise relevance propagation (LRP) to networks both with and without XAUG variables. The XAUG variables are concatenated with the intermediate layers after network-specific operations (such as convolution or recurrence), and used in the final layers of the network. The results of comparing networks with and without the addition of XAUG variables show that XAUG variables can be used to interpret classifier behavior, increase discrimination ability when combined with low-level features, and in some cases capture the behavior of the classifier completely. The LRP technique can be used to find relevant information the network is using, and when combined with the XAUG variables, can be used to rank features, allowing one to find a reduced set of features that capture part of the network performance. In the studies presented, adding XAUG variables to low-level DNNs increased the efficiency of classifiers by as much as 30-40%. In addition to performance improvements, an approach to quantify numerical uncertainties in the training of these DNNs is presented. 

Surface meteorological analyses are an essential input (termed “forcing”) for hydrologic modeling. This study investigated the sensitivity of different hydrologic model configurations to temporal variations of seven forcing variables (precipitation rate, air temperature, longwave radiation, specific humidity, shortwave radiation, wind speed, and air pressure). Specifically, the effects of temporally aggregating hourly forcings to hourly daily average forcings were examined. The analysis was based on 14 hydrological outputs from the Structure for Unifying Multiple Modeling Alternatives (SUMMA) model for the 671 Catchment Attributes and Meteorology for Large-Sample Studies (CAMELS) basins across the contiguous United States (CONUS). Results demonstrated that the hydrologic model sensitivity to temporally aggregating the forcing inputs varies across model output variables and model locations. We used Latin hypercube sampling to sample model parameters from eight combinations of three influential model physics choices (three model decisions with two options for each decision, i.e., eight model configurations). Results showed that the choice of model physics can change the relative influence of forcing on model outputs and the forcing importance may not be dependent on the parameter space. This allows for model output sensitivity to forcing aggregation to be tested prior to parameter calibration. More generally, this work provides a comprehensive analysis of the dependence of modeled outcomes on input forcing behavior, providing insight into the regional variability of forcing variable dominance on modeled outputs across CONUS.

 

Abstract Many measurements at the LHC require efficient identification of heavy-flavour jets, i.e. jets originating from bottom (b) or charm (c) quarks. An overview of the algorithms used to identify c jets is described and a novel method to calibrate them is presented. This new method adjusts the entire distributions of the outputs obtained when the algorithms are applied to jets of different flavours. It is based on an iterative approach exploiting three distinct control regions that are enriched with either b jets, c jets, or light-flavour and gluon jets. Results are presented in the form of correction factors evaluated using proton-proton collision data with an integrated luminosity of 41.5 fb -1 at  √s = 13 TeV, collected by the CMS experiment in 2017. The closure of the method is tested by applying the measured correction factors on simulated data sets and checking the agreement between the adjusted simulation and collision data. Furthermore, a validation is performed by testing the method on pseudodata, which emulate various mismodelling conditions. The calibrated results enable the use of the full distributions of heavy-flavour identification algorithm outputs, e.g. as inputs to machine-learning models. Thus, they are expected to increase the sensitivity of future physics analyses.