The Method of Uncertainty Minimization using Polynomial Chaos Expansions (MUM-PCE) was developed as a software tool to constrain physical models against experimental measurements. These models contain parameters that cannot be easily determined from first principles and so must be measured, and some which cannot even be easily measured. In such cases, the models are validated and tuned against a set of global experiments which may depend on the underlying physical parameters in a complex way. The measurement uncertainty will affect the uncertainty in the parameter values.

Included here are figures and relevant data for the work "Estimating Uncertainty in Robot Kinematics and Pose Measurements with Expectation-Maximization". We present a method to validate the measurement uncertainty of a metrology instrument without a priori estimates in the context of a kinematic calibration using Expectation-Maximization methods and extend our results to characterize post-calibration pose uncertainty for the manipulator throughout a workspace. This technique permits the robot kinematic model to be fitted simultaneously with a parameterized uncertainty model derived from direct-drive laser tracker kinematics. We demonstrate the performance of this algorithm in a simulated and experimental setting, achieving 6.4um position and 70.8 urad rotation error for kinematic calibration and statistically validating the fitted uncertainty model for points throughout the calibrated workspace.

Our goal is to explore the feasibility and usefulness of using a combination of covering arrays and machine learning models for predicting results of an agent- based simulation model within the vast parameter value combination space. The challenge is to select parameter values that are representative of the overall behavior of the model, so that we can train the machine learning model to be able to correctly predict behavior on previously untested areas of the parameter space. We have chosen Wilensky's Heat Bugs model in NetLogo for our study. It is a simple model, amenable to quick data generation, with a limited number of outputs to predict, and with emergent behavior. This model therefore allows exploration of this new approach.We utilize covering arrays to reduce the parameter value space systematically, run the model for each parameter set in the 2-way and 3-way covering arrays, train a random forest model on the 2-way data (33, 351 parameter combinations), and test its ability to predict the outcome of the simulation on the significantly larger 3-way data that was not seen during the training of the model (3, 971, 955 parameter combinations).

NeXLUncertainties.jl is a Julia package implementing algorithms for the propagation of uncertainties in multi-variate measurement models.

Figure summarizes uncertainty (error) in hourly gradient flux measurements by individual analyte. Flux uncertainty is derived from estimates of uncertainty in chemical gradients and turbulent transfer velocity. This dataset is associated with the following publication: Rumsey, I. Application of an online ion chromatography-based instrument for gradient flux measurements of speciated nitrogen and sulfur. ENVIRONMENTAL SCIENCE & TECHNOLOGY. American Chemical Society, Washington, DC, USA, 9(6): 2581-2592, (2016).

Our work towards a Structured Evaluation Methodology for Artificial Intelligence Technology (SEMAIT) aims to provide plots, tools, methods, and strategies to extract insights out of various machine learning (ML) and Artificial Intelligence (AI) data.Included in this software is the MIg analyZeR (mizr) R software package that produces various plots. It was initially developed within the Multimodal Information Group (MIG) at the National Institute of Standards and Technology (NIST).This software is documented, configured to be installed as an R package, and comes with an example SEMAIT script with an example (system, dataset, metrics, score) ML tuple set that we constructed ourselves.