HPC Ontology: Towards a Unified Ontology for Managing Training Datasets and AI Models for High-Performance Computing

Liao, Chunhua; Lin, Pei-Hung; Verma, Gaurav; Vanderbruggen, Tristan; Emani, Murali; Nan, Zifan; Shen, Xipeng

doi:10.1109/mlhpc54614.2021.00012

Cited by 12 publications

(7 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…• We also chose the Creative Commons 4.0 license (CC-BY 4.0) for the XPlacer dataset to fulfill the R1.1 principle. • We extended the HPC ontology [12] to provide required rich attributes to describe fine-grain data elements. A special unit ontology (QUDT) is also used to annotate the units for numerical values to enable maximum data interoperability and fulfill the RDA-I3-01M FAIRness indicator.…”

Section: Methodsmentioning

confidence: 99%

“…To address this problem, we are developing the HPC Ontology [12] to provide standard attributes which can be used to annotate fine-grain data elements in different subdomains of HPC, including GPU profiling results, program analysis (e. g. call graph analysis), and machine learning models (e. g. decision trees). The design of HPC Ontology is modular so it can be extended to include more subdomains in the future.…”

Section: B Improving Fairnessmentioning

confidence: 99%

See 1 more Smart Citation

Making Machine Learning Datasets and Models FAIR for HPC: A Methodology and Case Study

Lin

Liao

Chen

et al. 2022

2022 Fourth International Conference on Transdisciplinary AI (TransAI)

Self Cite

View full text Add to dashboard Cite

The FAIR Guiding Principles aim to improve the findability, accessibility, interoperability, and reusability of digital content by making them both human and machine actionable. However, these principles have not yet been broadly adopted in the domain of machine learning-based program analyses and optimizations for High-Performance Computing (HPC). In this paper, we design a methodology to make HPC datasets and machine learning models FAIR after investigating existing FAIRness assessment and improvement techniques. Our methodology includes a comprehensive, quantitative assessment for elected data, followed by concrete, actionable suggestions to improve FAIRness with respect to common issues related to persistent identifiers, rich metadata descriptions, license and provenance information. Moreover, we select a representative training dataset to evaluate our methodology. The experiment shows the methodology can effectively improve the dataset and model's FAIRness from an initial score of 19.1% to the final score of 83.0%.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: B Improving Fairnessmentioning

confidence: 99%

Making Machine Learning Datasets and Models FAIR for HPC: A Methodology and Case Study

Lin

Liao

Chen

et al. 2022

2022 Fourth International Conference on Transdisciplinary AI (TransAI)

Self Cite

View full text Add to dashboard Cite

show abstract

“…It relies on ontologies and ontology templates for task negotiation, data and ML feature annotation. (Liao et al, 2021) designed an ontology for high-performance computing (HPC) to make training datasets and AI models FAIR. Their ontology provides controlled vocabularies, explicit semantics, and formal knowledge representations.…”

Section: Literature Reviewmentioning

confidence: 99%

ISJM Volume 6, No.1 (2022)

2022

ISJ-Monte

View full text Add to dashboard Cite

In the last decade, the Albanian reader has welcomed the work of the Italian writer Elena Ferrante. The work that caused a stir and made her famous in Albania were the four volumes "The genious friend", which were successively published. What mostly attracts the Albanian reader is the reality the characters live in and the city of Naples where the events take place. Thanks to the way and style of writing, the Albanian reader experiences this reality because there he/she finds many things in common, almost the same mentality, not only the same lifestyle and habits, but also problems similar to those of southern Italy and especially Naples. The atavistic asphyxiation of the Neapolitan social classes only evokes ignorance and violence in daily life, and painlessly suppresses human intelligence. Elena, the main character of the work, embodies the characteristics of the city from which she wants to leave at all costs and where again the blood bond calls her and does not leave her free. Elena Ferrante portrays with a lot of elegance this reality which the Albanian reader can undoubtfully encounter in his/her daily life. Another motive that attracts the reader's attention is the close relationship of two friends, Lenu and Lila, a close friendship between girls, where love and envy, cooperation and rivalry, acceptance and rejection are the necessary directions of a society which would not exist, if it were in another, different form. For this and many other reasons the Neapolitan saga has been translated into 40 languages of the world and still continues to fascinate readers and publishers.

show abstract

“…Funded by DOE. This multi-institutional project aims to develop a generic High Performance Computing data management framework 9,10 to make both training data and AI models of scientific applications FAIR. • The FAIR Surrogate Benchmarks Initiative 11 .…”

Section: Fair Initiativesmentioning

confidence: 99%

FAIR for AI: An interdisciplinary, international, inclusive, and diverse community building perspective

Huerta¹,

Blaiszik²,

Brinson³

et al. 2022

Preprint

View full text Add to dashboard Cite

A foundational set of findable, accessible, interoperable, and reusable (FAIR) principles were proposed in 2016 as prerequisites for proper data management and stewardship, with the goal of enabling the reusability of scholarly data. The principles were also meant to apply to other digital assets, at a high level, and over time, the FAIR guiding principles have been reinterpreted or extended to include the software, tools, algorithms, and workflows that produce data. FAIR principles are now being adapted in the context of AI models and datasets. Here, we present the perspectives, vision, and experiences of researchers from different countries, disciplines, and backgrounds who are leading the definition and adoption of FAIR principles in their communities of practice, and discuss outcomes that may result from pursuing and incentivizing FAIR AI research. The material for this report builds on the FAIR for AI Workshop held at Argonne National Laboratory on June 7, 2022.

show abstract

HPC Ontology: Towards a Unified Ontology for Managing Training Datasets and AI Models for High-Performance Computing

Cited by 12 publications

References 38 publications

Making Machine Learning Datasets and Models FAIR for HPC: A Methodology and Case Study

Making Machine Learning Datasets and Models FAIR for HPC: A Methodology and Case Study

ISJM Volume 6, No.1 (2022)

FAIR for AI: An interdisciplinary, international, inclusive, and diverse community building perspective

Contact Info

Product

Resources

About