GitWorkflow for Active Learning: A Development Methodology Proposal for Data-Centric AI Projects

被引：0

作者：

Stieler, Fabian ^{[1
]}

Bauer, Bernhard ^{[1
]}

机构：

[1] Univ Augsburg, Inst Comp Sci, Augsburg, Germany

来源：

PROCEEDINGS OF THE 18TH INTERNATIONAL CONFERENCE ON EVALUATION OF NOVEL APPROACHES TO SOFTWARE ENGINEERING, ENASE 2023 | 2023年

关键词：

Active Learning; Software Engineering for Machine Learning; Machine Learning Operations;

D O I：

10.5220/0011988400003464

中图分类号：

TP31 [计算机软件];

学科分类号：

081202 ; 0835 ;

摘要：

As soon as Artificial Intelligence (AI) projects grow from small feasibility studies to mature projects, developers and data scientists face new challenges, such as collaboration with other developers, versioning data, or traceability of model metrics and other resulting artifacts. This paper suggests a data-centric AI project with an Active Learning (AL) loop from a developer perspective and presents "Git Workflow for AL": A methodology proposal to guide teams on how to structure a project and solve implementation challenges. We introduce principles for data, code, as well as automation, and present a new branching workflow. The evaluation shows that the proposed method is an enabler for fulfilling established best practices.

引用

页码：202 / 213

页数：12

共 50 条

[31] Taxonomy of machine learning paradigms: A data-centric perspective
Emmert-Streib, Frank
Dehmer, Matthias
[J]. WILEY INTERDISCIPLINARY REVIEWS-DATA MINING AND KNOWLEDGE DISCOVERY, 2022, 12 (05)
[32] Analyzing Data-Centric Properties for Graph Contrastive Learning
Trivedi, Puja
Lubana, Ekdeep Singh
Heimann, Mark
Koutra, Danai
Thiagarajan, Jayaraman J.
[J]. ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 35 (NEURIPS 2022), 2022,
[33] D-AI2-M: Ethanol Production Forecasting in Brazil Using Data-Centric Artificial Intelligence Methodology
Mello, Antonio
Giusti, Lucas
Tavares, Tarsila
Alexandrino, Fernando
Guedes, Gustavo
Soares, Jorge
Barbastefano, Rafael
Porto, Fabio
Carvalho, Diego
Ogasawara, Eduardo
[J]. IEEE Latin America Transactions, 2024, 22 (11): : 899 - 910
[34] wProjects: Data-centric Web Development for Female Nonprogrammers
Harshbarger, Nicole L.
Rosson, Mary Beth
[J]. 2012 IEEE SYMPOSIUM ON VISUAL LANGUAGES AND HUMAN-CENTRIC COMPUTING (VL/HCC), 2012, : 67 - 70
[35] IoT in the Fog: A Roadmap for Data-Centric IoT Development
Oteafy, Sharief M. A.
Hassanein, Hossam S.
[J]. IEEE COMMUNICATIONS MAGAZINE, 2018, 56 (03) : 157 - 163
[36] ydata-profiling: Accelerating data-centric AI with high-quality data
Clemente, Fabiana
Ribeiro, Goncalo Martins
Quemy, Alexandre
Santos, Miriam Seoane
Pereira, Ricardo Cardoso
Barros, Alex
[J]. NEUROCOMPUTING, 2023, 554
[37] A Reverse Data-Centric Process Design Methodology for Public Administration Processes
Kiss, Peter Jozsef
Klimko, Gabor
[J]. ELECTRONIC GOVERNMENT AND THE INFORMATION SYSTEMS PERSPECTIVE, EGOVIS 2019, 2019, 11709 : 85 - 99
[38] Reimagining Synthetic Tabular Data Generation through Data-Centric AI: A Comprehensive Benchmark
Hansen, Lasse
Seedat, Nabeel
van der Schaar, Mihaela
Petrovic, Andrija
[J]. ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 36 (NEURIPS 2023), 2023,
[39] A data-centric review of deep transfer learning with applications to text data
Bashath, Samar
Perera, Nadeesha
Tripathi, Shailesh
Manjang, Kalifa
Dehmer, Matthias
Streib, Frank Emmert
[J]. INFORMATION SCIENCES, 2022, 585 : 498 - 528
[40] Proposal of Data-Centric Network for Mobile and Dynamic Machine-to-Machine Communication
Matsubara, Daisuke
Yabusaki, Hitoshi
Okamoto, Satoru
Yamanaka, Naoaki
Takahashi, Tatsuro
[J]. IEICE TRANSACTIONS ON COMMUNICATIONS, 2013, E96B (11) : 2795 - 2806

← 1 2 3 4 5 →