Search NASA⌕ Search

SEARCH · Search NASA

Results for “Explanability”

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.

At least 37 records · Page 2

An alternative explanation of the change in T-dependence of the effective Debye-Waller factor at T{sub c} or T{sub B}

The cusp-like temperature dependence of the Debye-Waller factor or non-ergodicity parameter f{sub Q}(T) at some temperature T{sub c} above T{sub g} found by experiments in several fragile glassformers has been considered as critical evidence for validity of the ideal Mode Coupling Theory (MCT). A comprehensive review of experimental data of f{sub Q}(T) and beyond brings out various problems of the MCT predictions. For example, the molten salt, 0.4Ca(NO{sub 3}){sub 2}-0.6KNO{sub 3} (CKN), was the first glassformer measured by neutron scattering to verify the cusp-like behavior of f{sub Q}(T) at T{sub c} predicted by ideal MCT. While the fits of the other scaling laws of MCT to viscosity, light scattering, and dielectric relaxation data all give T{sub c} in the range from 368 to 375 K, there is no evidence of cusp-like behavior of f{sub Q}(T) at T{sub c} from more accurate neutron scattering data obtained later on by Mezei and Russina [J. Phys.: Condens. Matter 11, A341 (1999)] at temperatures below 400 K. In several molecular glass-formers, experiments have found at temperatures below T{sub c} that [1−f{sub Q}(T)] is manifested as nearly constant loss (NCL) in the frequency dependent susceptibility. The NCL persists down to below T{sub g} and is not predicted by the ideal MCT. No clear evidence of the change of T-dependence of f{sub Q}(T) at any T{sub c} was found in intermediate and strong glassformers, although ideal MCT does not distinguish fragile and strong glassformers in predicting the critical behavior of f{sub Q}(T) a priori. Experiments found f{sub Q}(T) changes T-dependence not only at T{sub c} but also at the glass transition temperature T{sub g}. The changes of T-dependence of f{sub Q}(T) at T{sub c} and T{sub g} are accompanied by corresponding changes of dynamic variables and thermodynamic quantities at T{sub B} ≈ T{sub c} and at T{sub g}. The dynamic variables include the relaxation time τ{sub α}(T), the non-exponentiality parameter n(T), and the generalized fragility m(T) of the structural α-relaxation. The thermodynamic quantities are the free volume deduced from positron annihilation spectroscopy, and the configurational entropy obtained from adiabatic calorimetry measurements. These changes of dynamic variables and thermodynamic quantities in temperature dependence at T{sub B} ≈ T{sub c} occur concurrently with the change of f{sub Q}(T) and suggest the effects are related, and have to be explained altogether. Since this task cannot be carried out by the ideal MCT, we have provided a different interpretation of f{sub Q}(T) and an alternative explanation of the change in its T-dependence of f{sub Q}(T) at T{sub B} ≈ T{sub c} as well as the other dynamic variables. We show f{sub Q}(T) originates from the dissipation of the molecules while caged by the anharmonic intermolecular potential, and manifested as the NCL at lower temperatures. The cusp-like change of T-dependence of f{sub Q}(T) at T{sub c} originates from the corresponding change of free volume and configurational entropy at T{sub B} ≈ T{sub c}, which also explains the simultaneous changes of the T-dependencies of the other dynamic variables. The alternative explanation is able to resolve the conundrum in CKN because T{sub B} is ≥400 K, and hence the change of T-dependence of f{sub Q}(T) at T{sub c} ≈ T{sub B} was not observed in data taken at temperatures lower than 400 K by Mezei and Russina. The alternative explanation also can rationalize the difference between fragile and non-fragile glassformers in the strength of the observed changes of f{sub Q}(T) at T{sub c} and T{sub g} as well as the other dynamic quantities at T{sub B} ≈ T{sub c} and T{sub g}.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Robust Explanations using Diverse Adversarially Trained Ensembles, Multi-Modal Contrastive Learning, and Attribution-based Confidence Metrics

The primary objective of this project is to strengthen the trustworthiness of AI systems by designing algorithms that make their internal decision-making processes more understandable to human users. This involves creating clear, interpretable explanations for AI decisions and developing metrics to assess these explanations' validity and reliability. Significant progress has been achieved through (i) developing symbolic explanations, (ii) generating meaningful interpretive insights, (iii) establishing accuracy and confidence metrics, and (iv) devising methods to evaluate the knowledge boundaries of AI models. To date, the research findings have been shared in peer-reviewed publications, with accompanying scientific and technical information (STI) detailed below.

97 MATHEMATICS AND COMPUTING↗

An Explanation of the Missing Flux from Boyajian's Mysterious Star

A previously unremarkable star in the constellation Cygnus has, in the past year, become known as the most mysterious object in our Galaxy. Boyajian’s star exhibits puzzling episodes of sporadic, deep dimming discovered in photometry with the Kepler Mission. Proposed explanations have focused on its obscuration by colliding exoplanets, exocomets, and even intervention of alien intelligence. These hypotheses have considered only phenomena external to the star because the radiative flux missing in the dimmings was believed to exceed the star’s storage capacity. We point out that modeling of variations in solar luminosity indicates that convective stars can store the required fluxes. It also suggests explanations for (a) a reported time-profile asymmetry of the short, deep dimmings and (b) a slower, decadal scale dimming reported from archival and Kepler photometry. Our findings suggest a broader range of explanations of Boyajian’s star that may produce new insights into stellar magneto-convection.

79 ASTRONOMY AND ASTROPHYSICS↗

Graphical explanation in an expert system for Space Station Freedom rack integration

The rationale and methodology used to incorporate graphics into explanations provided by an expert system for Space Station Freedom rack integration is examined. The rack integration task is typical of a class of constraint satisfaction problems for large programs where expertise from several areas is required. Graphically oriented approaches are used to explain the conclusions made by the system, the knowledge base content, and even at more abstract levels the control strategies employed by the system. The implemented architecture combines hypermedia and inference engine capabilities. The advantages of this architecture include: closer integration of user interface, explanation system, and knowledge base; the ability to embed links to deeper knowledge underlying the compiled knowledge used in the knowledge base; and allowing for more direct control of explanation depth and duration by the user. The graphical techniques employed range from simple statis presentation of schematics to dynamic creation of a series of pictures presented motion picture style. User models control the type, amount, and order of information presented.

Craig, F. G.↗

Explanations in knowledge systems - Design for Explainable Expert Systems

The Explainable Expert Systems framework (EES), in which the focus is on capturing those design aspects that are important for producing good explanations, including justifications of the system's actions, explications of general problem-solving strategies, and descriptions of the system's terminology, is discussed. EES was developed as part of the Strategic Computing Initiative of DARPA. Both the general principles from which the system was derived and how the system was derived from those principles can be represented in EES. The Program Enhancement Advisor, which is the main prototype on which the explanation work has been developed and tested, is presented. PEA is an advice system that helps users improve their Common Lisp programs by recommending transformations that enhance the user's code. How EES produces better explanations is shown.

Swartout, William↗

Explanation-based generalization of partially ordered plans

Most previous work in analytic generalization of plans dealt with totally ordered plans. These methods cannot be directly applied to generalizing partially ordered plans, since they do not capture all interactions among plan operators for all total orders of such plans. We introduce a new method for generalizing partially ordered plans. This method is based on providing explanation-based generalization (EBG) with explanations which systematically capture the interactions among plan operators for all the total orders of a partially-ordered plan. The explanations are based on the Modal Truth Criterion which states the necessary and sufficient conditions for ensuring the truth of a proposition at any point in a plan, for a class of partially ordered plans. The generalizations obtained by this method guarantee successful and interaction-free execution of any total order of the generalized plan. In addition, the systematic derivation of the generalization algorithms from the Modal Truth Criterion obviates the need for carrying out a separate formal proof of correctness of the EBG algorithms.

Kambhampati, Subbarao↗

Portfolios in Stochastic Local Search: Efficiently Computing Most Probable Explanations in Bayesian Networks

Portfolio methods support the combination of different algorithms and heuristics, including stochastic local search (SLS) heuristics, and have been identified as a promising approach to solve computationally hard problems. While successful in experiments, theoretical foundations and analytical results for portfolio-based SLS heuristics are less developed. This article aims to improve the understanding of the role of portfolios of heuristics in SLS. We emphasize the problem of computing most probable explanations (MPEs) in Bayesian networks (BNs). Algorithmically, we discuss a portfolio-based SLS algorithm for MPE computation, Stochastic Greedy Search (SGS). SGS supports the integration of different initialization operators (or initialization heuristics) and different search operators (greedy and noisy heuristics), thereby enabling new analytical and experimental results. Analytically, we introduce a novel Markov chain model tailored to portfolio-based SLS algorithms including SGS, thereby enabling us to analytically form expected hitting time results that explain empirical run time results. For a specific BN, we show the benefit of using a homogenous initialization portfolio. To further illustrate the portfolio approach, we consider novel additive search heuristics for handling determinism in the form of zero entries in conditional probability tables in BNs. Our additive approach adds rather than multiplies probabilities when computing the utility of an explanation. We motivate the additive measure by studying the dramatic impact of zero entries in conditional probability tables on the number of zero-probability explanations, which again complicates the search process. We consider the relationship between MAXSAT and MPE, and show that additive utility (or gain) is a generalization, to the probabilistic setting, of MAXSAT utility (or gain) used in the celebrated GSAT and WalkSAT algorithms and their descendants. Utilizing our Markov chain framework, we show that expected hitting time is a rational function - i.e. a ratio of two polynomials - of the probability of applying an additive search operator. Experimentally, we report on synthetically generated BNs as well as BNs from applications, and compare SGSs performance to that of Hugin, which performs BN inference by compilation to and propagation in clique trees. On synthetic networks, SGS speeds up computation by approximately two orders of magnitude compared to Hugin. In application networks, our approach is highly competitive in Bayesian networks with a high degree of determinism. In addition to showing that stochastic local search can be competitive with clique tree clustering, our empirical results provide an improved understanding of the circumstances under which portfolio-based SLS outperforms clique tree clustering and vice versa.

Mengshoel, Ole J.↗

GRIP: Constraint-based Explanation of Missing Answers for Graph Queries

Abstract: A useful feature in graph query engines is to clarify “Why certain entities (nodes, attribute values or edges) are missing” in query answers. This task is even more challenging when the relevant data is already missing in the underlying data source. Missing data, on the other hand, can be inferred by enforcing data constraints for graphs. We demonstrate GRIP, a system that exploits data constraints to clarify missing answers for graph queries. (1) Constraint-based ex- planation. Given a desired yet missing entity in the query answer, GRIP ensures to generate finite and minimal sequences of data con- strains (an “explanation”) that should be consecutively enforced to ?? to ensure its occurrence for the same query. (2) Answering “why” and“how” questions. Users can query GRIP with both“Why”(“Why” the element is missing) and “How” questions (“How” to refine the graph to include the missing answer). GRIP engine supports run- time generation of explanations by incrementally maintaining a set of bi-directional search trees. (3) Interactive exploration. GRIP provides a user-friendly GUI to support interactive ad visual exploration of explanations, including both automated generation and step-by-step inspection of graph manipulations.

graphs↗

Investigation of rotating mode behavior in BWR out-of-phase limit cycle oscillations – Part 2: TRACE/PARCS model and physical explanation

Previous neutronic/thermal-hydraulic (TH) coupled numerical simulations using full-core TRACE/PARCS and SIMULATE-3K boiling water reactor (BWR) models have shown evidence of a specific “rotating mode” behavior (steady rotation of the symmetry line, i.e. constant phase shift of approximately 90° between the first two azimuthal modes) in BWR out-of-phase limit cycle oscillations, regardless of initial conditions and even if the first two azimuthal modes have different natural frequencies. This suggests a nonlinear coupling between these modes; otherwise, the phase shift between these modes would change at a constant rate during the limit cycle. The previous paper (“Part 1”) presented a series of results to examine this rotating behavior with a reduced-order model. The goal of the present study is to provide additional analyses of the predicted rotating mode behavior using higher-fidelity numerical modeling, as well as a physical explanation for why this mode is favored over side-to-side or other oscillatory behaviors from a TH perspective. Results are presented using TRACE and TRACE/PARCS for a small number of parallel channels, which confirmed that the conclusions developed from the reduced-order model remain applicable when applying a full two-fluid, six-equation, finite-volume modeling approach. From these results, a physical explanation has been put forth to explain why the rotating symmetry line behavior is preferred from a TH standpoint, demonstrating that predominantly out-of-phase unstable systems are most unstable when the variation in the total inlet flow rate is minimized (which minimizes the effective single-phase to two-phase pressure drop ratio) and that the rotating mode is the most successful in minimizing this total flow rate variation as compared with the side-to-side case or any other oscillation pattern. The conclusion is that the rotating mode will be favored for any out-of-phase unstable system of parallel channels with no neutronic feedback or relatively weak neutronic feedback. Here, previous analyses have indicated that systems with sufficiently strong neutronic coupling may favor the side-to-side oscillation mode over the rotating mode; this topic is left as a subject of future investigation.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

Explanation of the seasonal variation of cosmic multiple muon events observed with the NOvA Near Detector

The flux of cosmic ray muons at the Earth’s surface exhibits seasonal variations due to changes in the temperature of the atmosphere affecting the production and decay of mesons in the upper atmosphere. Using 1473 live days of data collected by the NuMI Off-axis 𝜈 𝑒 Appearance (NOvA) Near Detector during 2018–2022, we studied the seasonal pattern in the multiple-muon event rate. The data confirm an anticorrelation between the multiple-muon event rate and effective atmospheric temperature, consistent across all the years of data. Previous analyses from MINOS and NOvA saw a similar anticorrelation but did not include an explanation. We find that this anticorrelation is driven by altitude–geometry effects as the average muon production height changes with the season. This has been studied with a CORSIKA cosmic ray simulation package by varying atmospheric parameters, and provides an explanation to a longstanding discrepancy between the seasonal phases of single and multiple-muon events.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Nonresonant explanation for the Y(4260) structure observed in the e{sup +}e{sup -}{yields}J/{psi}{pi}{sup +}{pi}{sup -} process

In this work, we propose a nonresonant explanation for the Y(4260) structure observed in the e{sup +}e{sup -}{yields}J/{psi}{pi}{sup +}{pi}{sup -} process, i.e., Y(4260) is not a genuine resonance. Our result indicates that the Y(4260) structure can be reproduced by the interference of production amplitudes of the e{sup +}e{sup -}{yields}J/{psi}{pi}{sup +}{pi}{sup -} processes via direct e{sup +}e{sup -} annihilation and through intermediate charmonia {psi}(4160)/{psi}(4415). Besides describing the Y(4260) line shape in e{sup +}e{sup -}{yields}J/{psi}{pi}{sup +}{pi}{sup -} well, such a nonresonant explanation for the Y(4260) structure naturally answers why there is no evidence of Y(4260) in the exclusive open-charm decay channel and R-value scan.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

A Workflow for Visual Diagnostics of Binary Classifiers using Instance-Level Explanations

In this paper, we propose a visual analytics workflow to help data scientists explore, diagnose, and understand the decisions made by a binary classifier. The approach leverages “instance-level explanations”, measures of local feature relevance that explain single instances and uses them to build a set of visual representations that guide the users in their investigation. The workflow is based on three main visual representations and steps: one based on aggregate statistics to see how data distributes across correct or incorrect decisions; one based on explanations to understand which features are used to make these decisions; and one based on raw data, to derive insights on potential root causes for the observed patterns. The work is validated through a long-term collaboration with a group of healthcare professionals who used our method to make sense of machine learning models they developed. The case studies from this collaboration demonstrate that the proposed method helps experts derive useful knowledge about the model and the phenomena it describes and also generates useful hypotheses on how a model can be improved.

Krause, Josua↗

Interpreting Black-Box Classifiers Using Instance-Level Visual Explanations

To realize the full potential of machine learning in diverse real- world domains, it is necessary for model predictions to be readily interpretable and actionable for the human in the loop. Analysts, who are the users but not the developers of machine learning models, often do not trust a model because of the lack of transparency in associating predictions with the underlying data space. To address this problem, we propose Rivelo, a visual analytic interface that enables analysts to understand the causes behind predictions of binary classifiers by interactively exploring a set of instance-level explanations. These explanations are model-agnostic, treating a model as a black box, and they help analysts in interactively probing the high-dimensional binary data space for detecting features relevant to predictions. We demonstrate the utility of the interface with a case study analyzing a random forest model on the sentiment of Yelp reviews about doctors.

Tamagnini, Paolo↗

Connections between Student Explanations and Arguments from Evidence about Plant Growth

We investigate how students connect explanations and arguments from evidence about plant growth and metabolism—two key practices described by the Next Generation Science Standards. This study reports analyses of interviews with 22 middle and high school students postinstruction, focusing on how their sense-making strategies led them to interpret—or misinterpret—scientific explanations and arguments from evidence. The principles of conservation of matter and energy can provide a framework for making sense of phenomena, but our results show that some students reasoned about plant growth as an action enabled by water, air, sunlight, and soil rather than a process of matter and energy transformation. These students reinterpreted the hypotheses and results of standard investigations of plant growth, such as van Helmont's experiment, to match their own understanding of how plants grow. Only the more advanced students consistently interpreted mass changes in plants or soil as evidence of movement of matter. We also observed that a higher degree of scaffolding during some of the interview questions allowed mid-level students to improve their responses. We describe our progress and challenges developing teaching materials with scaffolding to improve students’ understanding of plant growth and metabolism.

54 ENVIRONMENTAL SCIENCES↗

Integrating adaptive learning with post hoc model explanation and symbolic regression to build interpretable surrogate models

Abstract We develop a materials informatics workflow to build an interpretable surrogate model for micromagnetic simulations. Our goal is to predict the energy barrier of a moving isolated skyrmion in rare-earth-free $$\hbox {Mn}_4$$ Mn 4 N. Our approach integrates adaptive learning with post hoc model explanation and symbolic regression methods. We discuss an unexplored acquisition function (information condensing active learning) within the adaptive learning loop and compare it with the known standard deviation function for efficient navigation of the search space. Model-agnostic post hoc explanation techniques then uncover trends learned by the trained model, which we then leverage to constrain the expressions used for symbolic regression. Graphical abstract

Biswas, Ankita↗

Io's sodium cloud - Explanation of the east-west asymmetries

An explanation of the east-west asymmetries of Io's sodium clouds is suggested based upon the force experienced by sodium atoms as they resonantly scatter sunlight and upon the ability of this force to alter the cloud atom orbits, as determined by the gravitational fields of Io and Jupiter. Orbit calculations illustrating the time-dependent changes in the sodium atom trajectories produced by the solar radiation force as a function of Io's orbital location are presented in support of this explanation. This theory agrees with the idea that sodium atoms are emitted from the satellite nonisotropically and at a constant rate. These conclusions may be important in studying the global effects of active volcanoes and the intense satellite plasma torus discovered for Io from Voyager I spacecraft measurements.

Smyth, W. H.↗

CLIPS enhanced with objects, backward chaining and explanation facilities

In this project we extend C Language Production System (CLIPS), an existing Expert System shell, by creating three new options. Specifically, first we create a compatible with CLIPS environment that allows for defining objects and object hierarchies, second we provide means to implement backward chaining in a pure forward chaining environment, and finally we give some simple explanation facilities for the derivations the system has made. Objects and object hierarchies are extended so that facts can be automatically inferred, and placed in the fact base. Backward chaining is implemented by creating run time data structures which hold the derivation process allowing for a depth first search. The backward chaining mechanism works not only with ground facts, but also creates bindings for every query that involves variables, and returns the truth value of such a query as well as the relevant variable bindings. Finally, the WHY and HOW explanation facilities allow for a complete examination of the derivation process, the rules triggered, and the bindings created. The entire system is integrated with the original CLIPS code, and all of its routines can be invoked as CLIPS commands.

Aldrobi, M.↗

Initialization and Restart in Stochastic Local Search: Computing a Most Probable Explanation in Bayesian Networks

For hard computational problems, stochastic local search has proven to be a competitive approach to finding optimal or approximately optimal problem solutions. Two key research questions for stochastic local search algorithms are: Which algorithms are effective for initialization? When should the search process be restarted? In the present work we investigate these research questions in the context of approximate computation of most probable explanations (MPEs) in Bayesian networks (BNs). We introduce a novel approach, based on the Viterbi algorithm, to explanation initialization in BNs. While the Viterbi algorithm works on sequences and trees, our approach works on BNs with arbitrary topologies. We also give a novel formalization of stochastic local search, with focus on initialization and restart, using probability theory and mixture models. Experimentally, we apply our methods to the problem of MPE computation, using a stochastic local search algorithm known as Stochastic Greedy Search. By carefully optimizing both initialization and restart, we reduce the MPE search time for application BNs by several orders of magnitude compared to using uniform at random initialization without restart. On several BNs from applications, the performance of Stochastic Greedy Search is competitive with clique tree clustering, a state-of-the-art exact algorithm used for MPE computation in BNs.

Mengshoel, Ole J.↗