Search NASA⌕ Search

SEARCH · Search NASA

Results for “Large language model”

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 595 records · Page 33

EHR-BERT: A BERT-based model for effective anomaly detection in electronic health records

Objective: Physicians and clinicians rely on data contained in electronic health records (EHRs), as recorded by health information technology (HIT), to make informed decisions about their patients. The reliability of HIT systems in this regard is critical to patient safety. Consequently, better tools are needed to monitor the performance of HIT systems for potential hazards that could compromise the collected EHRs, which in turn could affect patient safety. In this paper, we propose a new framework for detecting anomalies in EHRs using sequence of clinical events. This new framework, EHR-Bidirectional Encoder Representations from Transformers (BERT), is motivated by the gaps in the existing deep-learning related methods, including high false negatives, sub-optimal accuracy, higher computational cost, and the risk of information loss. EHR-BERT is an innovative framework rooted in the BERT architecture, meticulously tailored to navigate the hurdles in the contemporary BERT method; thus, enhancing anomaly detection in EHRs for healthcare applications.Methods: The EHR-BERT framework was designed using the Sequential Masked Token Prediction (SMTP) method. This approach treats EHRs as natural language sentences and iteratively masks input tokens during both training and prediction stages. This method facilitates the learning of EHR sequence patterns in both directions for each event and identifies anomalies based on deviations from the normal execution models trained on EHR sequences.Results: Extensive experiments on large EHR datasets across various medical domains demonstrate that EHR-BERT markedly improves upon existing models. It significantly reduces the number of false positives and enhances the detection rate, thus bolstering the reliability of anomaly detection in electronic health records. This improvement is attributed to the model’s ability to minimize information loss and maximize data utilization effectively.Conclusion: EHR-BERT showcases immense potential in decreasing medical errors related to anomalous clinical events, positioning itself as an indispensable asset for enhancing patient safety and the overall standard of healthcare services. The framework effectively overcomes the drawbacks of earlier models, making it a promising solution for healthcare professionals to ensure the reliability and quality of health data.

96 KNOWLEDGE MANAGEMENT AND PRESERVATION↗

Improving the Capabilities and Computational Efficiency of the RTE+RRTMGP Radiation Code (Final Report)

This report details progress on the RTE+RRTMGP radiation codes made during the period of performance. RTE+RRTMGP is a set of codes for computing radiative fluxes in planetary atmospheres. RRTMGP uses a k-distribution to provide an optical description (absorption and possibly Rayleigh optical depth) of the gaseous atmosphere, along with the relevant source functions, on a pre-determined spectral grid given temperatures, pressures, and gas concentration. RTE computes fluxes given spectrally-resolved optical descriptions and source functions. Spectrally-resolved fluxes are summarized (“reduced”) via a user extensible class. The initial release of the code and the design choices are described in Pincus et al. 2019; the codes are available on Github. Although RRTMGP was based on current (at the time) empirical spectroscopic data, RTE and RRTMGP were developed in large part to modernize software practices. The design focused on flexibility broadly interpreted: by separating code from data and allowing data to drive computation; in coupling to the host model (e.g. the coupling of clouds to radiative fluxes is user-controlled); with respect to programming languages (computational tasks are accessed via widely-compatible C interfaces); and with respect to hardware (the codes run on a range of CPU and GPU architectures). The code also puts an emphasis on modularity and clarity. RTE+RRTMGP v1.0 was released in September 20219. This award supported the evolution of the RTE+RRTMGP code base to support greater flexibility, accuracy, and efficiency.

54 ENVIRONMENTAL SCIENCES↗

HVAC and Control Templates for the Modelica Buildings Library

This article reports on our experience in creating Modelica models for systems with thousands of configurations and closed-loop controls. The development of such templates required exploration of class parameterization techniques and data structures for handling large sets of equipment parameters. By describing these issues and the approach taken, we show how the Modelica language can support advanced templating logic. The main limitation we encountered relates to parameter assignment and propagation. The interpretation of parameter attributes at user interface runtime, or the handling of non-trivial constructs involving record classes at compile time is not consistently supported by Modelica tools. This leads to choices that are difficult to make when looking for a generic implementation.

Gautier, Antoine↗

A Vehicle Management End-to-End Testing and Analysis Platform for Validation of Mission and Fault Management Algorithms to Reduce Risk for NASA's Space Launch System

The engineering development of the new Space Launch System (SLS) launch vehicle requires cross discipline teams with extensive knowledge of launch vehicle subsystems, information theory, and autonomous algorithms dealing with all operations from pre-launch through on orbit operations. The characteristics of these spacecraft systems must be matched with the autonomous algorithm monitoring and mitigation capabilities for accurate control and response to abnormal conditions throughout all vehicle mission flight phases, including precipitating safing actions and crew aborts. This presents a large and complex system engineering challenge, which is being addressed in part by focusing on the specific subsystems involved in the handling of off-nominal mission and fault tolerance with response management. Using traditional model based system and software engineering design principles from the Unified Modeling Language (UML) and Systems Modeling Language (SysML), the Mission and Fault Management (M&FM) algorithms for the vehicle are crafted and vetted in specialized Integrated Development Teams (IDTs) composed of multiple development disciplines such as Systems Engineering (SE), Flight Software (FSW), Safety and Mission Assurance (S&MA) and the major subsystems and vehicle elements such as Main Propulsion Systems (MPS), boosters, avionics, Guidance, Navigation, and Control (GNC), Thrust Vector Control (TVC), and liquid engines. These model based algorithms and their development lifecycle from inception through Flight Software certification are an important focus of this development effort to further insure reliable detection and response to off-nominal vehicle states during all phases of vehicle operation from pre-launch through end of flight. NASA formed a dedicated M&FM team for addressing fault management early in the development lifecycle for the SLS initiative. As part of the development of the M&FM capabilities, this team has developed a dedicated testbed that integrates specific M&FM algorithms, specialized nominal and off-nominal test cases, and vendor-supplied physics-based launch vehicle subsystem models. Additionally, the team has developed processes for implementing and validating these algorithms for concept validation and risk reduction for the SLS program. The flexibility of the Vehicle Management End-to-end Testbed (VMET) enables thorough testing of the M&FM algorithms by providing configurable suites of both nominal and off-nominal test cases to validate the developed algorithms utilizing actual subsystem models such as MPS. The intent of VMET is to validate the M&FM algorithms and substantiate them with performance baselines for each of the target vehicle subsystems in an independent platform exterior to the flight software development infrastructure and its related testing entities. In any software development process there is inherent risk in the interpretation and implementation of concepts into software through requirements and test cases into flight software compounded with potential human errors throughout the development lifecycle. Risk reduction is addressed by the M&FM analysis group working with other organizations such as S&MA, Structures and Environments, GNC, Orion, the Crew Office, Flight Operations, and Ground Operations by assessing performance of the M&FM algorithms in terms of their ability to reduce Loss of Mission and Loss of Crew probabilities. In addition, through state machine and diagnostic modeling, analysis efforts investigate a broader suite of failure effects and associated detection and responses that can be tested in VMET to ensure that failures can be detected, and confirm that responses do not create additional risks or cause undesired states through interactive dynamic effects with other algorithms and systems. VMET further contributes to risk reduction by prototyping and exercising the M&FM algorithms early in their implementation and without any inherent hindrances such as meeting FSW processor scheduling constraints due to their target platform - ARINC 653 partitioned OS, resource limitations, and other factors related to integration with other subsystems not directly involved with M&FM such as telemetry packing and processing. The baseline plan for use of VMET encompasses testing the original M&FM algorithms coded in the same C++ language and state machine architectural concepts as that used by Flight Software. This enables the development of performance standards and test cases to characterize the M&FM algorithms and sets a benchmark from which to measure the effectiveness of M&FM algorithms performance in the FSW development and test processes.

Trevino, Luis↗

National Ignition Facility Project Completion and Control System Status

The National Ignition Facility (NIF) is the world's largest and most energetic laser experimental system providing a scientific center to study inertial confinement fusion (ICF) and matter at extreme energy densities and pressures. Completed in 2009, NIF is a stadium-sized facility containing a 1.8-MJ, 500-TW 192-beam ultraviolet laser and target chamber. A cryogenic tritium target system and suite of optical, X-ray and nuclear diagnostics will support experiments in a strategy to achieve fusion ignition starting in 2010. Automatic control of NIF is performed by the large-scale Integrated Computer Control System (ICCS), which is implemented by 2 MSLOC of Java and Ada running on 1300 front-end processors and servers. The ICCS framework uses CORBA distribution for interoperation between heterogeneous languages and computers. Laser setup is guided by a physics model and shots are coordinated by data-driven distributed workflow engines. The NIF information system includes operational tools and a peta-scale repository for provisioning experimental results. This paper discusses results achieved and the effort now underway to conduct full-scale operations and prepare for ignition.

42 ENGINEERING↗

Quantifying Groundwater Response and Uncertainty in Beaver‐Influenced Mountainous Floodplains Using Machine Learning‐Based Model Calibration

Abstract Beavers ( Castor canadensis ) alter river corridor hydrology by creating ponds and inundating floodplains, and thereby improving surface water storage. However, the impact of inundation on groundwater, particularly in mountainous alluvial floodplains with permeable gravel/cobble layers overlain by a soil layer, remains uncertain. Numerical modeling across various floodplain structures considers topographic and sediment complexity and multidirectional flow, linking inundation to groundwater response. This study develops a model‐data integration workflow to address uncertainty in groundwater response to beaver‐induced inundations in a mountainous alluvial floodplain in the Upper Colorado River Basin. Uncertain factors include seasonal hydrologic dynamics, hydraulic conductivities, floodplain structures, and meteorological forcings. We employed an ensemble of groundwater models, based on geophysical and hydrologic data, with machine learning‐based calibration using a neural density estimator. This allowed us to quantify the vertical flux from the soil layer to the permeable gravel bed, the down‐valley underflow within the gravel bed, and their ratios. Results show a significant increase in the vertical flux relative to down‐valley underflow, from 2 during dry pond periods to 20 during wet periods, serving as an analogy for conditions without and with beaver ponds. The study highlights the influence of floodplain structure on groundwater storage, water balance, and water quality impacted by beaver ponds. A thick gravel bed layer, with a large down‐valley underflow, minimizes the effect of beaver‐induced inundation on water quality. We emphasize the need for field‐scale measurements of floodplain structure and improved characterization of evapotranspiration changes to reduce uncertainty in groundwater response. Plain Language Summary Beavers change the flow of water in river corridors by creating ponds, expanding wetlands, and flooding floodplains. This increases surface water area, promotes plant growth, and enhances biodiversity. However, the impact of this flooding on groundwater flow is not well understood, especially in mountainous areas with gravel layers where water moves easily beneath soil. In this study, we used numerical modeling to investigate how beaver ponds influence groundwater in a mountainous floodplain of the Upper Colorado River Basin. We adapted a machine learning method to validate our numerical models using multiple field data sets. Our findings show that beaver ponds significantly increase vertical water flow from the soil to the gravel during wet periods, compared to when the ponds are fully drained. The study also highlights the importance of floodplain structure in controlling both water flow in gravel layers along the river direction and vertical flow from the soil to the gravel with the presence of beavers. To reduce uncertainty in groundwater response, we emphasize the need for more field‐scale measurements of floodplain structure, hydraulic properties, and evapotranspiration changes. Key Points Floodplain structures and hydraulic conductivities are important for groundwater response with beaver ponds in mountainous floodplains Large down‐valley underflow in permeability‐stratified floodplains reduces beaver‐induced impacts on groundwater storage and water quality Machine learning‐based model calibration methods are effective for estimating posterior distributions of groundwater model parameters

Wang, Lijing↗

A Vehicle Management End-to-End Testing and Analysis Platform for Validation of Mission and Fault Management Algorithms to Reduce Risk for NASAs Space Launch System

The engineering development of the National Aeronautics and Space Administration's (NASA) new Space Launch System (SLS) requires cross discipline teams with extensive knowledge of launch vehicle subsystems, information theory, and autonomous algorithms dealing with all operations from pre-launch through on orbit operations. The nominal and off-nominal characteristics of SLS's elements and subsystems must be understood and matched with the autonomous algorithm monitoring and mitigation capabilities for accurate control and response to abnormal conditions throughout all vehicle mission flight phases, including precipitating safing actions and crew aborts. This presents a large and complex systems engineering challenge, which is being addressed in part by focusing on the specific subsystems involved in the handling of off-nominal mission and fault tolerance with response management. Using traditional model-based system and software engineering design principles from the Unified Modeling Language (UML) and Systems Modeling Language (SysML), the Mission and Fault Management (M&FM) algorithms for the vehicle are crafted and vetted in Integrated Development Teams (IDTs) composed of multiple development disciplines such as Systems Engineering (SE), Flight Software (FSW), Safety and Mission Assurance (S&MA) and the major subsystems and vehicle elements such as Main Propulsion Systems (MPS), boosters, avionics, Guidance, Navigation, and Control (GNC), Thrust Vector Control (TVC), and liquid engines. These model-based algorithms and their development lifecycle from inception through FSW certification are an important focus of SLS's development effort to further ensure reliable detection and response to off-nominal vehicle states during all phases of vehicle operation from pre-launch through end of flight. To test and validate these M&FM algorithms a dedicated test-bed was developed for full Vehicle Management End-to-End Testing (VMET). For addressing fault management (FM) early in the development lifecycle for the SLS program, NASA formed the M&FM team as part of the Integrated Systems Health Management and Automation Branch under the Spacecraft Vehicle Systems Department at the Marshall Space Flight Center (MSFC). To support the development of the FM algorithms, the VMET developed by the M&FM team provides the ability to integrate the algorithms, perform test cases, and integrate vendor-supplied physics-based launch vehicle (LV) subsystem models. Additionally, the team has developed processes for implementing and validating the M&FM algorithms for concept validation and risk reduction. The flexibility of the VMET capabilities enables thorough testing of the M&FM algorithms by providing configurable suites of both nominal and off-nominal test cases to validate the developed algorithms utilizing actual subsystem models such as MPS, GNC, and others. One of the principal functions of VMET is to validate the M&FM algorithms and substantiate them with performance baselines for each of the target vehicle subsystems in an independent platform exterior to the flight software test and validation processes. In any software development process there is inherent risk in the interpretation and implementation of concepts from requirements and test cases into flight software compounded with potential human errors throughout the development and regression testing lifecycle. Risk reduction is addressed by the M&FM group but in particular by the Analysis Team working with other organizations such as S&MA, Structures and Environments, GNC, Orion, Crew Office, Flight Operations, and Ground Operations by assessing performance of the M&FM algorithms in terms of their ability to reduce Loss of Mission (LOM) and Loss of Crew (LOC) probabilities. In addition, through state machine and diagnostic modeling, analysis efforts investigate a broader suite of failure effects and associated detection and responses to be tested in VMET to ensure reliable failure detection, and confirm responses do not create additional risks or cause undesired states through interactive dynamic effects with other algorithms and systems. VMET further contributes to risk reduction by prototyping and exercising the M&FM algorithms early in their implementation and without any inherent hindrances such as meeting FSW processor scheduling constraints due to their target platform - the ARINC 6535-partitioned Operating System, resource limitations, and other factors related to integration with other subsystems not directly involved with M&FM such as telemetry packing and processing. The baseline plan for use of VMET encompasses testing the original M&FM algorithms coded in the same C++ language and state machine architectural concepts as that used by FSW. This enables the development of performance standards and test cases to characterize the M&FM algorithms and sets a benchmark from which to measure their effectiveness and performance in the exterior FSW development and test processes. This paper is outlined in a systematic fashion analogous to a lifecycle process flow for engineering development of algorithms into software and testing. Section I describes the NASA SLS M&FM context, presenting the current infrastructure, leading principles, methods, and participants. Section II defines the testing philosophy of the M&FM algorithms as related to VMET followed by section III, which presents the modeling methods of the algorithms to be tested and validated in VMET. Its details are then further presented in section IV followed by Section V presenting integration, test status, and state analysis. Finally, section VI addresses the summary and forward directions followed by the appendices presenting relevant information on terminology and documentation.

Trevino, Luis↗

Open Innovation for a NASA Architecture Library

NASA’s Center of Excellence for Collaborative Innovation (CoECI) uses open innovation, or “crowdsourcing”, to access the global public to find ideas, concepts, designs, or solutions that meet a previously unmet need possibly resulting in significant advances in performance. The Center of Excellence for Collaborative Innovation was launched at the request of the White House Office of Science and Technology Policy. This is both a non-traditional method of innovation and a non-traditional method of outreach to the public to involve them in space technologies and programs. It has been used often for software development and new hardware technology. In this case we applied it to innovate with systems engineering tools for creating space architectures. The challenge was sponsored by NASA Engineering and Safety Center Systems Engineering Technical Fellow as part of a program for NASA’s adoption of MBSE. It was a trial to see if there would be as much participation or quality submissions with this more specialized topic and skill. The challenge sought space architecture representations and decompositions to create a library of modeled parts in a system modeling language (SysML). Mission architects mostly start from scratch to build model elements representing the functional and physical architecture of a system in SysML. There are a few beginning libraries, but these are also local to a program or group. A common library will save system engineers a large amount of time, will allow project stakeholders to recognize common graphics and quickly understand the architecture options. The challenge was promoted internationally, especially through professional organizations and universities with a systems engineering focus. It was open for 4 months, purposefully over the winter holiday break time to allow participants extra time outside of work or school. The challenge was designed so that expertise in space hardware was not necessary but getting to play with models of space architecture could provide motivation to participate. We did not receive as many entries as other broader outreach challenges, but the ones we received were extremely thorough and high quality. Solutions came from individuals and teams, students and professional consultants from the United States and Europe. We learned a few lessons about how to engage with the public and what characteristics of a problem result in good crowdsourcing results. The outreach challenge produced several useful ideas and modeled space elements, and the group will be engaging the winners to learn more about their new approaches.

innovation↗

Studying language evolution in the age of big data

The increasing availability of large digital corpora of cross-linguistic data is revolutionizing many branches of linguistics. Overall, it has triggered a shift of attention from detailed questions about individual features to more global patterns amenable to rigorous, but statistical, analyses. This engenders an approach based on successive approximations where models with simplified assumptions result in frameworks that can then be systematically refined, always keeping explicit the methodological commitments and the assumed prior knowledge. Therefore, they can resolve disputes between competing frameworks quantitatively by separating the support provided by the data from the underlying assumptions. These methods, though, often appear as a ‘black box’ to traditional practitioners. In fact, the switch to a statistical view complicates comparison of the results from these newer methods with traditional understanding, sometimes leading to misinterpretation and overly broad claims. We describe here this evolving methodological shift, attributed to the advent of big, but often incomplete and poorly curated, data, emphasizing the underlying similarity of the newer quantitative to the traditional comparative methods and discussing when and to what extent the former have advantages over the latter. In this review, we cover briefly both randomization tests for detecting patterns in a largely model-independent fashion and phylolinguistic methods for a more model-based analysis of these patterns. We foresee a fruitful division of labor between the ability to computationally process large volumes of data and the trained linguistic insight declaring worthy prior commitments and interesting hypotheses in need of comparison.

96 KNOWLEDGE MANAGEMENT AND PRESERVATION↗

Automated information retrieval using CLIPS

Expert systems have considerable potential to assist computer users in managing the large volume of information available to them. One possible use of an expert system is to model the information retrieval interests of a human user and then make recommendations to the user as to articles of interest. At Cal Poly, a prototype expert system written in the C Language Integrated Production System (CLIPS) serves as an Automated Information Retrieval System (AIRS). AIRS monitors a user's reading preferences, develops a profile of the user, and then evaluates items returned from the information base. When prompted by the user, AIRS returns a list of items of interest to the user. In order to minimize the impact on system resources, AIRS is designed to run in the background during periods of light system use.

Raines, Rodney Doyle, III↗

Scaling Irregular Applications through Data Aggregation and Software Multithreading

Bioinformatics, data analytics, semantic databases, knowledge discovery are emerging high performance application areas that exploit dynamic, linked data structures such as graphs, unbalanced trees or unstructured grids. These data structures usually are very large, requiring significantly more memory than available on single shared memory systems. Additionally, these data structures are difficult to partition on distributed memory systems. They also present poor spatial and temporal locality, thus generating unpredictable memory and network accesses. The Partitioned Global Address Space (PGAS) programming model seems suitable for these applications, because it allows using a shared memory abstraction across distributed-memory clusters. However, current PGAS languages and libraries are built to target regular remote data accesses and block transfers. Furthermore, they usually rely on the Single Program Multiple Data (SPMD) parallel control model, which is not well suited to the fine grained, dynamic and unbalanced parallelism of irregular applications. In this paper we present {\bf GMT} (Global Memory and Threading library), a custom runtime library that enables efficient execution of irregular applications on commodity clusters. GMT integrates a PGAS data substrate with simple fork/join parallelism and provides automatic load balancing on a per node basis. It implements multi-level aggregation and lightweight multithreading to maximize memory and network bandwidth with fine-grained data accesses and tolerate long data access latencies. A key innovation in the GMT runtime is its thread specialization (workers, helpers and communication threads) that realize the overall functionality. We compare our approach with other PGAS models, such as UPC running using GASNet, and hand-optimized MPI code on a set of typical large-scale irregular applications, demonstrating speedups of an order of magnitude.

Aggregation↗

Machine Learning Approaches to Predicting Induced Seismicity and Imaging Geothermal Reservoir Properties

This project developed machine learning (ML) methods, lab data sets, and field data to advance geothermal exploration and geothermal energy production. The work had three focus areas. One involved the development of ML methods to use microearthquakes (MEQs) for imaging geothermal reservoir properties and improving subsurface characterization – most importantly the evolution of permeability within the evolving reservoir. This part of the work included development of ML approaches for automated MEQ location, focal mechanism determination and identification of earthquake precursors. The second area focused on using MEQ signals generated by geothermal exploration and production to predict the relationship between fluid injection and seismicity. Here, we extended to reservoir scale our success in using ML to predict laboratory earthquakes and fault zone stress state. The third focus area was on lab experiments. Here, we developed new ML models for lab earthquake prediction and identification of precursors to failure to improve earthquake forecasting and early warning in geothermal settings. Major outcomes of our work include ML models that learn from MEQ signals during geothermal exploration and production to predict induced seismicity. MEQs occur naturally in connection with drilling and energy production. We developed ML methods to use the seismic waves from these events to characterize the elastic, hydraulic and poromechanical properties of reservoirs. Our work illuminated fracture geometry and the evolution of fracture permeability by incorporating seismic coda wave analysis and ML methods to relate fluid injection and seismicity. We significantly expanded laboratory earthquake prediction to include methods that use both passive measurements of microearthquakes within the lab fault zones and also active source acoustic measurements of fault zone elastic properties. These methods can now predict fault zone stress state, time to failure and the magnitude of lab earthquakes. Our work showed that repetitive stick- slip failure events during frictional sliding (the lab equivalent of earthquakes) are preceded by a cascade of micro-failure events that radiate energy in a manner that foretells unstable failure – manifest as laboratory MEQs. We documented a mapping between fracture properties and statistical attributes of elastic radiation. We extended existing works to geothermal reservoir scale and developed ML methods to determine reservoir permeability, fracture properties, and their evolution during geothermal energy production. An attractive feature of ML algorithms is their ability to handle big datasets and reveal patterns and correlations that may remain invisible to conventional analyses. Our work connected data from field, laboratory and intermediate scales to study permeability, stress, strength, fracture stiffness and geometry. At the field scale we used data from the Newberry Volcano field site, UtahFORGE, EGS Collab, and also the Bedretto underground research lab in Switzerland. These data sets are bridging the gap between the lab scale, theory, and reservoir scale. Our work produced plain language summaries to improve public understanding of DOE research. We also developed openly distributed ML and seismicity datasets for use by all researchers and we published connections between induced seismicity in geothermal areas and reservoir properties including permeability, fracture properties, and stress state. Our models are designed for the large data sets of induced seismicity typically associated with geothermal sites. We produced labeled event catalogs and used them on geothermal data to assess how ML can facilitate geothermal production and exploration. All datasets are available on the GDR Productivity: The project produced 32 publications in peer reviewed journals (two are in review). It supported the work of 6 PhD students, 40 conference presentations, 6 keynote talks at national meetings, and mentoring and professional development for 4 postdoctoral fellows.

15 GEOTHERMAL ENERGY↗

Reconfigurable middleware architectures for large scale sensor networks

Wireless sensor networks, in an e ffort to be energy efficient, typically lack the high-level abstractions of advanced programming languages. Though strong, the dichotomy between these two paradigms can be overcome. The SENSIX software framework, described in this dissertation, uniquely integrates constraint-dominated wireless sensor networks with the flexibility of object-oriented programming models, without violating the principles of either. Though these two computing paradigms are contradictory in many ways, SENSIX bridges them to yield a dynamic middleware abstraction unifying low-level resource-aware task recon figuration and high-level object recomposition.

99 GENERAL AND MISCELLANEOUS↗

Robotic Planning under Uncertainty in Spatiotemporal Environments in Expeditionary Science

In the expeditionary sciences, spatiotemporally varying environments -- hydrothermal plumes, algal blooms, lava flows, or animal migrations -- are ubiquitous. Mobile robots are uniquely well-suited to study these dynamic, mesoscale natural environments. We formalize expeditionary science as a sequential decision-making problem, modeled using the language of partially-observable Markov decision processes (POMDPs). Solving the expeditionary science POMDP under real-world constraints requires efficient probabilistic modeling and decision-making in problems with complex dynamics and observational models. Previous work in informative path planning, adaptive sampling, and experimental design have shown compelling results, largely in static environments, using data-driven models and information-based rewards. However, these methodologies do not trivially extend to expeditionary science in spatiotemporal environments: they generally do not make use of scientific knowledge such as equations of state dynamics, they focus on information gathering as opposed to scientific task execution, and they make use of decision-making approaches that scale poorly to large, continuous problems with long planning horizons and real-time operational constraints. In this work, we discuss these and other challenges related to probabilistic modeling and decision-making in expeditionary science, and present some of our preliminary work that addresses these gaps. We ground our results in a real expeditionary science deployment of an autonomous underwater vehicle (AUV) in the deep ocean for hydrothermal vent discovery and characterization. Our concluding thoughts highlight remaining work to be done, and the challenges that merit consideration by the reinforcement learning and decision-making community.

Preston, Victoria↗

Automatic specification of reliability models for fault-tolerant computers

The calculation of reliability measures using Markov models is required for life-critical processor-memory-switch structures that have standby redundancy or that are subject to transient or intermittent faults or repair. The task of specifying these models is tedious and prone to human error because of the large number of states and transitions required in any reasonable system. Therefore, model specification is a major analysis bottleneck, and model verification is a major validation problem. The general unfamiliarity of computer architects with Markov modeling techniques further increases the necessity of automating the model specification. Automation requires a general system description language (SDL). For practicality, this SDL should also provide a high level of abstraction and be easy to learn and use. The first attempt to define and implement an SDL with those characteristics is presented. A program named Automated Reliability Modeling (ARM) was constructed as a research vehicle. The ARM program uses a graphical interface as its SDL, and it outputs a Markov reliability model specification formulated for direct use by programs that generate and evaluate the model.

Liceaga, Carlos A.↗

Improving the Performance of NEML2 with Modern Graph Compilation Backends

NEML2 vectorizes constitutive-model evaluation for large-scale multiphysics simulation, using PyTorch as its tensor backend so that a batch of material-point updates runs on CPU or GPU through a single implementation. In the two prior reports in this series it was a C++-native library, deployed through TorchScript tracing and just-in-time (JIT) compilation; it has since been rewritten from the ground up into a Python-native library deployed through Ahead-of-Time Inductor (AOTInductor), a modern PyTorch graph-compilation backend. The rewrite is driven by a persistent tension, not a language preference: NEML2 composes constitutive models at runtime from a registry of small, independently-authored pieces, and that flexibility is difficult to reconcile with the compile-time knowledge an efficient GPU kernel needs. This report documents the rewrite and the investment that accompanied it: the AOTInductor export pipeline that turns a Python-authored model into a portable, Python-free compiled artifact loadable from pure C++; the eager and compiled runtimes and the new implicit solver layer built on them; a head-to-head benchmark of legacy JIT against AOTInductor; the physics-model catalog and its worked examples; the developer tooling; and the corresponding overhaul of MOOSE’s NEML2 integration that lets MOOSE consume it. A central objective is to examine whether modern PyTorch graph-compilation backends are effective for MOOSE GPU integration. The benchmark answers directly: AOTInductor outperforms legacy JIT on every GPU scenario measured, by 1.0–4.5×. Modern graph-compilation backends are effective for MOOSE GPU integration, and AOTInductor specifically – not compilation in the abstract – is why.

Hu, Gary (Tianchen) [Argonne National Laboratory (↗

Compiler Optimization for Extreme-Scale Scripting

The data-driven task parallelism execution model can support parallel programming models that are well suited for large-scale distributed-memory parallel computing, for example, simulations and analysis pipelines running on clusters and clouds. We describe a novel compiler intermediate representation and optimizations for this execution model, including adaptions of standard techniques alongside novel techniques. These techniques are applied to Swift/T, a high-level scripting language for flexible data flow composition of functions, which may be serial or use lower-level parallel programming models such as MPI and OpenMP. This paper presents preliminary results, indicating that our compiler optimizations reduce communication overhead by 70% to 93% on distributed-memory systems.

Armstrong, Timothy G.↗

Data Quality Monitoring for the Hadron Calorimeters Using Transfer Learning for Anomaly Detection

The proliferation of sensors brings an immense volume of spatio-temporal (ST) data in many domains, including monitoring, diagnostics, and prognostics applications. Data curation is a time-consuming process for a large volume of data, making it challenging and expensive to deploy data analytics platforms in new environments. Transfer learning (TL) mechanisms promise to mitigate data sparsity and model complexity by utilizing pre-trained models for a new task. Despite the triumph of TL in fields like computer vision and natural language processing, efforts on complex ST models for anomaly detection (AD) applications are limited. In this study, we present the potential of TL within the context of high-dimensional ST AD with a hybrid autoencoder architecture, incorporating convolutional, graph, and recurrent neural networks. Motivated by the need for improved model accuracy and robustness, particularly in scenarios with limited training data on systems with thousands of sensors, this research investigates the transferability of models trained on different sections of the Hadron Calorimeter of the Compact Muon Solenoid experiment at CERN. The key contributions of the study include exploring TL’s potential and limitations within the context of encoder and decoder networks, revealing insights into model initialization and training configurations that enhance performance while substantially reducing trainable parameters and mitigating data contamination effects.

47 OTHER INSTRUMENTATION↗