SEARCH · Search NASA
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.
Develop the next generation GEOS model that enables an increase in resolution and scalability to meet future NASA Global Modeling and Assimilation Office (GMAO) requirements.
Goal : Develop the next generation GEOS model that enables an increase in resolution and scalability to meet future NASA Global Modeling and Assimilation Office (GMAO) requirements.
Language models are fast approaching human-like understanding of natural language. They have been shown to perform equally, if not better than humans in a myriad of language tasks such as next sentence prediction, question answering, entity extraction etc. Part of the success of the models are owed to the fact that they have been trained on varied natural language text over the internet. By virtue of this, the models do not contain the semantic information present in Earth science literature. Hence, there is a lot of room for improvement when using these models for earth science specific tasks. In this work, we showcase our approach on developing Earth science specific language models. Furthermore, we justify the need for such a model by using the embeddings generated by the model to perform a domain specific downstream task that performs better than a generic model.
This tutorial explores the design and implementation issues arising in the development of domain-specific languages for trace analysis. It introduces the audience to the general concepts underlying such special-purpose languages building upon the authors' own experiences in developing both external domain specific languages and systems, such as EAGLE, HAWK, RULER and LOGSCOPE, and the more recent internal domain-specific language and system TRACECONTRACT within the SCALA language.
SLEEC (Semantics-rich Libraries for Effective Exascale Computation) was a project funded by the Department of Energy X-Stack Program, award number DE-SC0008629. The initial project period was September 2012–August 2015. The project was renewed for an additional year, expiring August 2016. Finally, the project received a no-cost extension, leading to a final expiry date of August 2017. Modern applications, especially those intended to run at exascale, are not written from scratch. Instead, they are built by stitching together various carefully-written, hand-tuned libraries. Correctly composing these libraries is difficult, but traditional compilers are unable to effectively analyze and transform across abstraction layers. Domain specific compilers integrate semantic knowledge into compilers, allowing them to transform applications that use particular domain-specific languages, or domain libraries. But they do not help when new domains are developed, or applications span multiple domains. SLEEC aims to fix these problems. To do so, we are building generic compiler and runtime infrastructures that are semantics-aware but not domain-specific. By performing optimizations related to the semantics of a domain library, the same infrastructure can be made generic and apply across multiple domains.
The QCOR project consists of 4 main research tasks: 1) scientific applications and algorithm development for machine learning (ML) and quantum theories of condensed matter systems and lattice gauge theories; 2) high-level quantum-classical (HLQC) language and domain specific language (DSL) development; 3) compiler and interfaces development for the HLQC language; and 4) hardware-level control, optimization, and algorithms libraries. In the succeeding sections we report on progress towards our research tasks during the 2020 calendar year.
The paper provides an introduction to the track: “Program- ming - What is Next?”, organized by the authors as part of ISoLA 2021: the 9th International Symposium On Leveraging Applications of Formal Methods, Verification and Validation. A total of 14 papers were pre- sented in the track, with responses to the question: what are the trends in current more recent programming languages, and what can be expected of future languages?. The track covers such topics as general-purpose programming languages, domain-specific languages, formal methods and modeling languages, textual versus graphical languages, and application programming versus embedded programming.
In the aviation domain, there are many applications for machine learning and artificial intelligence tools that utilize natural language. For example, there is a desire to know the commonalities in written safety reports such as voluntary post incidents reports or aerial wildfire operations reports to better understand the risks present. Another use-case is the possibility of extracting airspace procedures and constraints currently written in documents such as Letters of Agreement. These applications can benefit from the use of state-of-the-art natural language processing techniques when adapted to the language/phraseology specific to the aviation domain. This paper evaluates the viability of adaptation of NLP tools to the aviation domain by fine-tuning transformer based models using aviation data sets. In 2018, a novel language model based on neural units (also called transformers) was created and became known as “Bidirectional Encoder Representations from Transformers” or BERT. This architecture combined with large amounts of English training data and innovative semi-supervised training tasks set the standard for what would later emerge as Large Language Models. The performance of these models was further improved by hyperparameter tuning and refinement of the semi-supervised training task and resulted in “Robustly Optimized BERT Pre-training Approach through hyperparameter tuning” or RoBERTa models. These pre-trained Large Language Models proved to be useful for a wide variety of natural language processing tasks such as text classification and question answering through a process called fine-tuning. The transformer architecture with pre-trained weights served as the basis with the last few layers replaced with layers fine-tuned to perform a new task e.g., a layer that provides a label for the entire input text. This process of fine-tuning can also be used to adapt the models to new domains; e.g., BioBERT started with the pre-trained BERT model and was completed by additional fine-tuning and training on biomedical documents. Transformer-based architectures can also be used to create rich representations of text called embeddings which can serve as the input to other machine learning models. This allows simpler algorithms such as logistic regression to use context-rich representations of the text while still remaining quick to train and evaluate. In the world of aviation, there is a growing demand for natural language processing and understanding but the domain presents unique challenges. Due to the technical content (and specialized language) of most aviation documents, fine-tuning pre-trained Large Language Models to specific tasks has not met the benchmark on natural language processing tasks set by simpler models trained from scratch on the data. To address this deficiency, this paper evaluates the improvements from fine-tuning a Large Language Model on a large set of aviation documents using the original semi-supervised training tasks before performing specific natural language tasks. In fine-tuning, a domain-specific dataset is used on the original training task but with the pre-trained Large Language Model instead of starting from a random initialization. This approach allows the model to be adapted to the specific domain language without discarding the information gained from training on general English data. This paper utilized two major dataset types to train and assess the RoBERTa fine-tuning performance. The first are 7,057 Letters of Agreement which are Federal Aviation Administration (FAA) documents that formalize airspace operations across the national airspace system. They contain many examples of ‘aviation English’ using domain specific terminology and phrasing which serves as a representative basis to perform the semi-supervised fine-tuning. The second type is the 494 document classification labels to be used for evaluation. This down-stream evaluation aims to show the performance of the fine-tuned model, better understand how much data is needed for an effective fine-tuning, and how fine-tuning can be adapted for different applications in-the domain. After semi-supervised training, evaluation begins by encoding the documents for classification using the fine-tuned RoBERTa model. Then a logistic regression classifier is trained to label the document type and compared against our ground truth labels. This currently leads to a 82.8% accuracy on 10-fold cross validation showing improvement over baseline RoBERTa which achieved 81.0%. We plan to measure the improvements on additional tasks and it is expected that these improvements will lead to more robust models that can tackle the natural language processing challenges present in aviation datasets.
The software currently compiles LLVM IR into Solidity (Ethereum’s dominant programming language) using LLVM’s pass library. Specifically, his compiler allows us to convert an arbitrary DSL into Solidity. We focus specifically on converting Domain Specific Languages into Solidity due to their ease of use, and provable properties. By creating a toolchain to compile lightweight domain-specific languages into Ethereum's dominant language, Solidity, we allow non-specialists to effectively develop safe and useful smart contracts. For example lawyers from a certain firm can have a proprietary DSL that codifies basic laws safely converted to Solidity to be securely executed on the blockchain. In another example, a simple provenance tracking language can be compiled and securely executed on the blockchain.
An analysis description language is a domain specific language capable of describing the contents of an LHC analysis in a standard and unambiguous way, independent of any computing framework. It is designed for use by anyone with an interest in, and knowledge of, LHC physics, i.e., experimentalists, phenomenologists and other enthusiasts. Adopting analysis description languages would bring numerous benefits for the LHC experimental and phenomenological communities ranging from analysis preservation beyond the lifetimes of experiments or analysis software to facilitating the abstraction, design, visualization, validation, combination, reproduction, interpretation and overall communication of the analysis contents. Here, we introduce the analysis description language concept and summarize the current efforts ongoing to develop such languages and tools to use them in LHC analyses.
In this paper we describe TRACECONTRACT, an API for trace analysis, implemented in the SCALA programming language. We argue that for certain forms of trace analysis the best weapon is a high level programming language augmented with constructs for temporal reasoning. A trace is a sequence of events, which may for example be generated by a running program, instrumented appropriately to generate events. The API supports writing properties in a notation that combines an advanced form of data parameterized state machines with temporal logic. The implementation utilizes SCALA's support for defining internal Domain Specific Languages (DSLs). Furthermore SCALA's combination of object oriented and functional programming features, including partial functions and pattern matching, makes it an ideal host language for such an API.
Explore the source record for details and available documents.
RuleBuilder is a tool for drawing graphs represented by the BioNetGen rule-based modeling language (BNGL), a domain-specific language used to formulate mathematical models of biochemical interaction networks. This framework provides an intuitive plain-text representation of such networks using a graph-based formalism. The interactions between the model's components are graph-rewriting rules that denote the minimum context needed for the interaction to occur. In some cases the context required for reactions to occur can grow quite large, and the lengthy textual description of the interaction can be difficult to interpret or debug. RuleBuilder is designed to make this process simpler through visualization of individual rule or pattern strings. The software assists in the reading of existing models by converting the BNGL strings of interest into a graph-based representation composed of nodes and edges. RuleBuilder also enables the user to build the visual representation directly from its interface, composing the BNGL string as components and operators are added to the graphical object.
Chemist is a domain-specific language targeting the QC domain. Chemist has been developed focusing on performance and user-friendliness. Using Chemist, QC tasks are defined using familiar domain concepts such as molecules, wave functions, and operators. The domain objects are hierarchical to ensure a systematic encapsulation of information. Key features of Chemist include: extensibility, the ability to alias existing data, and the ability to succinctly define many common QC tasks.
Ultimately, scientific numerical models need quantified output uncertainties so that modeling can evolve to better match reality. Documenting model input uncertainties and variabilities is a necessary first step toward that goal. Without known input parameter uncertainties, model sensitivities are all one can determine, and without code verification, output uncertainties are simply not reliable. The basic premise of uncertainty markup is to craft a tolerance and tagging mini-language that offers a natural, unobtrusive presentation and does not depend on parsing each type of input file format. Each file is marked up with tolerances and optionally, associated tags that serve to label the parameters and their uncertainties. The evolution of such a language, often called a Domain Specific Language or DSL, is given in [1], but in final form it parallels tolerances specified on an engineering drawing, e.g., 1 +/- 0.5, 5 +/- 10%, 2 +/- 10 where % signifies percent and o signifies order of magnitude. Tags, necessary for error propagation, can be added by placing a quotation-mark-delimited tag after the tolerance, e.g., 0.7 +/- 20% 'T_effective'. In addition, tolerances might have different underlying distributions, e.g., Uniform, Normal, or Triangular, or the tolerances may merely be intervals due to lack of knowledge (uncertainty). Finally, to address pragmatic considerations such as older models that require specific number-field formats, C-style format specifiers can be appended to the tolerance like so, 1.35 +/- 10U_3.2f. As an example of use, consider figure 1, where a chemical reaction input file is has been marked up to include tolerances and tags per table 1. Not only does the technique provide a natural method of specifying tolerances, but it also servers as in situ documentation of model uncertainties. This tolerance language comes with a utility to strip the tolerances (and tags), to provide a path to the nominal model parameter file. And, as shown in [1], having the ability to quickly mark and identify model parameter uncertainties facilitates error propagation, which in turn yield output uncertainties.
The field of runtime verification has during the last decade seen a multitude of systems for monitoring event sequences (traces) emitted by a running system. The objective is to ensure correctness of a system by checking its execution traces against formal specifications representing requirements. A special challenge is data parameterized events, where monitors have to keep track of the combination of control states as well as data constraints, relating events and the data they carry across time points. This poses a challenge wrt. efficiency of monitors, as well as expressiveness of logics. Data automata is a form of automata where states are parameterized with data, supporting monitoring of data parameterized events. We describe the full details of a very simple API in the Scala programming language, an internal DSL (Domain-Specific Language), implementing data automata. The small implementation suggests a design pattern. Data automata allow transition conditions to refer to other states than the source state, and allow target states of transitions to be inlined, offering a temporal logic flavored notation. An embedding of a logic in a high-level language like Scala in addition allows monitors to be programmed using all of Scala's language constructs, offering the full flexibility of a programming language. The framework is demonstrated on an XML processing scenario previously addressed in related work.
The Fortran mimetic abstraction language ("Formal") is a domain-specific language (DSL) embedded in Fortran 202Y [1]. Formal provides novel software abstractions for simulating phenomena governed by the partial differential equations (PDEs) of vector and tensor calculus. Such equations model an extremely broad set of physical phenomena, ranging from atmospheric winds to light propagation. Formal's data structures and algorithms mimic in form and behavior continuous functions and operators. Formal supports these mathematical constructs using mimetic discretizations that define a discrete calculus satisfying various tensor calculus theorems, thereby ensuring high-fidelity representations of the physics being modeled. [2] Formal 0.1.0 also lays a foundation for the future use of Fortran 202Y type-safe templates to facilitate the formal verification of tensor contractions in computational physics and artificial intelligence [3]. [1] "Fortran 202Y" is Fortran standard committee's informal designation for the next Fortran revision, which will likely be "Fortran 2028". [2] Corbino, J. and Castillo, J. (2020) Journal of Computational and Applied Mathematics, https://doi.org/10.1016/j.cam.2019.06.042. [3] Haveraaen, M., Järvi, J., & Rouson, D. (2019). Reflecting on Generics for Fortran. https://j3-fortran.org/doc/year/19/19-188.pdf.