An interpretable and scalable model for rapid flood extent forecasting using satellite imagery and machine learning with rotated EOF analysis
Explore the source record for details and available documents.
Hung-Hsien Wan↗
SEARCH · Search NASA
Results for “Model interpretation”
Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
Planet Four: Derived South Polar Martian Winds Interpreted Using Mesoscale Modeling
Explore the source record for details and available documents.
Ganna Portyankina↗
Calibrating Healthcare AI: Towards Reliable and Interpretable Deep Predictive Models
Explore the source record for details and available documents.
Computer science↗
Using Direct Error Predictors to Improve Model Safety and Interpretability
Explore the source record for details and available documents.
Computer science↗
Learning to Branch with Interpretable Machine Learning Models
The data consists of a slide deck that was presented at the INFORMS 2023 conference. The presentation summarizes our approach to learning how to branch and compares our approach to the popular solver SCIP and a state-of-the-art ML-based branching rule.
Bayramoglu, Selin↗
Interpretation of pilot opinion by application of multiloop models to a VTOL flight simulator task
Mathematical model for interpreting pilot opinion and selection of optimum control sensitivity for VTOL aircraft hovering task
Miller, D. P.↗
Interpretation of borehole strain measurements using surrogate modeling-based optimization
Interpreting strain data measured during well testing requires inverting poroelastic forward models set up to represent an aquifer or reservoir. One approach is to use stochastic methods to conduct the inversion. Subsurface parameters such as elastic modulus, permeability, and geometry of heterogeneities are estimated by searching the parameter space. This is feasible, but cumbersome, requiring more than a week of computation using many hundreds of computer nodes in one of our earlier analyses. This motivated us to consider alternative methods, including an artificial neural network (ANN) as a surrogate model.
Roudini, Soheil↗
Towards Trustworthy and Interpretable Deep Learning-assisted Ecohydrological Models
The transformational science question we plan to address is: How do we leverage in-situ observations and simulations from process-based ecohydrological model to construct interpretable and trustworthy deep learning (DL) models for improved reliability of prediction of quantities of interest (QoIs) under hydro-climatic extremes?
54 ENVIRONMENTAL SCIENCES↗
NLIZE: A Perturbation-Driven Visual Interrogation Tool for Analyzing and Interpreting Natural Language Inference Models
With the recent advances in deep learning, neural network models have obtained state-of-the-art performances for many linguistic tasks in natural language processing. However, this rapid progress also brings enormous challenges. The opaque nature of a neural network model leads to hard-to-debug-systems and difficult-to-interpret mechanisms. Here, we introduce a visualization system that, through a tight yet flexible integration between visualization elements and the underlying model, allows a user to interrogate the model by perturbing the input, internal state, and prediction while observing changes in other parts of the pipeline. We use the natural language inference problem to illustrate how a perturbation-driven paradigm can help domain experts assess the potential limitation of a model, probe its inner states, and interpret and form hypotheses about fundamental model mechanisms such as attention.
97 MATHEMATICS AND COMPUTING↗
Studies of Trace Gas Chemical Cycles Using Observations, Inverse Methods and Global Chemical Transport Models
For interpreting observational data, and in particular for use in inverse methods, accurate and realistic chemical transport models are essential. Toward this end we have, in recent years, helped develop and utilize a number of three-dimensional models including the Model for Atmospheric Transport and Chemistry (MATCH).
Prinn, Ronald G.↗
AttentionFire_v1.0: interpretable machine learning fire model for burned-area predictions over tropics
Abstract. African and South American (ASA) wildfires account for more than 70 % of global burned areas and have strong connection to local climate for sub-seasonal to seasonal wildfire dynamics. However, representation of the wildfire–climate relationship remains challenging due to spatiotemporally heterogenous responses of wildfires to climate variability and human influences. Here, we developed an interpretable machine learning (ML) fire model (AttentionFire_v1.0) to resolve the complex controls of climate and human activities on burned areas and to better predict burned areas over ASA regions. Our ML fire model substantially improved predictability of burned areas for both spatial and temporal dynamics compared with five commonly used machine learning models. More importantly, the model revealed strong time-lagged control from climate wetness on the burned areas. The model also predicted that, under a high-emission future climate scenario, the recently observed declines in burned area will reverse in South America in the near future due to climate changes. Our study provides a reliable and interpretable fire model and highlights the importance of lagged wildfire–climate relationships in historical and future predictions.
58 GEOSCIENCES↗
Collaborative Research: Enabling multi-scale studies of magnetic reconnection with interpretable data-driven models
The development of accurate reduced descriptions and improved closures for magnetic reconnection is an important and a long‐standing challenge in plasma physics. The four‐fluid approach, and associated closures, that were investigated have the potential to improve the accuracy of plasma fluid models, capturing physical effects which would otherwise require a kinetic description. If successful, this approach could have an important impact for the modeling of laboratory and space plasmas. The major goals of this project were to develop new machine learning (ML) tools based on sparse and symbolic regression techniques, and to extract interpretable and generalizable reduced models (e.g., in the form of partial differential equations - PDEs) from data generated by first principles plasma simulations. Preserving interpretability of such data‐driven models is key to addressing the long‐standing theoretical and numerical challenges. Prior proof‐of‐principle studies have demonstrated the enormous potential of this approach, by recovering the well‐established hierarchy of plasma equations (from Vlasov to MHD) from data produced by particle‐in‐cell (PIC) simulations. Our goal in this project was to extend and apply these new tools to construct better kinetic closures for magnetic reconnection; to derive better models of particle injection and acceleration by this fundamental plasma process; and to use this understanding to accelerate the development of multi‐scale plasma algorithms. While our immediate focus was on the problem of magnetic reconnection, the tools that were will developed are general and applicable to other areas of plasma physics, and more broadly to many‐body phenomena. We anticipate that the development of these multi‐scale models will have a significant impact across different areas of plasma science, from fusion to space and astrophysical plasmas.
70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
Nonlinear ARMA models for the D(st) index and their physical interpretation
Time series models successfully reproduce or predict geomagnetic activity indices from solar wind parameters. A method is presented that converts a type of nonlinear filter, the nonlinear Autoregressive Moving Average (ARMA) model to the nonlinear damped oscillator physical model. The oscillator parameters, the growth and decay, the oscillation frequencies and the coupling strength to the input are derived from the filter coefficients. Mathematical methods are derived to obtain unique and consistent filter coefficients while keeping the prediction error low. These methods are applied to an oscillator model for the Dst geomagnetic index driven by the solar wind input. A data set is examined in two ways: the model parameters are calculated as averages over short time intervals, and a nonlinear ARMA model is calculated and the model parameters are derived as a function of the phase space.
Vassiliadis, D.↗
ERBE bidirectional model consistency check
A short analysis is presented of Earth Radiation Budget Experiment (ERBE) errors inherent in the directional models used for data interpretation. The models were all developed on the basis of experience with the Nimbus-7 ERB experiment, which had a spatial resolution one-third that of ERBE instrumentation. A pseudo-directional model is defined to simulate the ERBE scanner data, using the assumptions that the average radiant exitance for any particular scene is independent of the viewing geometry, geographic location and time the data is collected. The directionality of the view angle and solar zenith angle is accounted for by a method of bins.
Baldwin, D. G.↗
Numerical Simulation of a Chemically Reacting Sorbent Bed for LSS Applications
A detailed numerical model of a chemisorption bed has been developed. The model is based on the constant pressure mass transport equation for gaseous flow through a packed bed, and the equation for diffusion and reaction within a spherical particle. Because there is a wealth of data from the NASA and the Navy bodies of literature, the LiOH-H2O-CO2 system is chosen for application of the model and interpretation of results. Prior models of this system from the life support literature are limited. The current model incorporates many of the features of elaborate models developed for investigation of industrial systems or energy applications (e.g., coal, desulphurization): it distinguishes bulk convection and bed dispersion; mass transport to the particle surface, transport within the particle, and reaction. It uses the nonsteady (not pseudo-steady state) form of the equations. The chemistry is modeled as a multi-step, reversible reaction with evolving solid structure. The resulting system of equations is large. The ODEPACK family of solvers is used to integrate the system. Reaction coefficients are determined by experiment. Typical results of the model are illustrated with mission input parameters. Using the model, an explanation is offered for 1) the varied performance results found after pre-breathing (or after simulated pre-breathe conditions), 2) interrupted use and 3) low temperature use. In addition, options for a reusable canister are explored. The computational resource implications of adding energy equations are discussed briefly, as are applicability to other relevant space and undersea systems.
Luna, Bernadette↗
An interpretable and transferrable vision transformer model for rapid materials spectra classification
An interpretable and transferrable Vision Transformer (ViT) model was developed for classifying individual materials from their XRD and FTIR spectra.
Chen, Zhenru↗
Demystifying Cyberattacks: Potential for Securing Energy Systems With Explainable AI : Preprint
Modernization of energy systems has led to in- creased interactions among multiple critical infrastructures and diverse stakeholders making the challenge of operational decision making more complex and at times beyond cognitive capabilities of human operators. The state-of-the-art machine learning and deep learning approaches show promise of supporting users with complex decision-making challenges, such as those occurring in our rapidly transforming cyber-physical energy systems. However, successful adoption of data-driven decision support technology for critical infrastructure will be dependent on the ability of these technologies to be trustworthy and contextually interpretable. In this paper, we investigate the feasibility of implementing XAI for interpretable detection of cyberattacks in the energy system. Leveraging a proof-of-concept simulation use case of detection of a data falsification attack on a photovoltaic system using XGBoost algorithm, we demonstrate how Local Interpretable Model-Agnostic Explanations (LIME), a flavor XAI approach, can help provide contextual and actionable interpretation of cyberattack detection.
artificial intelligence↗
A model of electronic map interpretation
This paper describes an experiment that provides data for the development of a cognitive model of pilot flight navigation. The model views navigation as a process involving the alignment of mental images with the perceptual view out of the cockpit. The data support a three stage model: (1) the perceptual encoding of the map display, (2) mental rotation of the mental image, and (3) comparison of the image to the environment. The variables that significantly influence the processes embodied in the model in decreasing importance are: speed of processing, display sequencing, map complexity, and rotation angle of the map. The model can be used as a preliminary computational tool in predicting the navigational component of pilot situational awareness.
Aretz, Anthony J.↗