A Model-Based Approach to Experiment-Driven Evolution of ML Workflows

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216208%3A11320%2F25%3A10508014" target="_blank" >RIV/00216208:11320/25:10508014 - isvavai.cz</a>

Výsledek na webu

<a href="https://doi.org/10.5220/0013380500003896" target="_blank" >https://doi.org/10.5220/0013380500003896</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.5220/0013380500003896" target="_blank" >10.5220/0013380500003896</a>

Jazyk výsledku

angličtina

Název v původním jazyce

A Model-Based Approach to Experiment-Driven Evolution of ML Workflows

Popis výsledku v původním jazyce

Machine Learning (ML) has advanced significantly, yet the development of ML workflows still relies heavily on expert intuition, limiting standardization. MLOps integrates ML workflows for reliability, while AutoML automates tasks like hyperparameter tuning. However, these approaches often overlook the iterative and experimental nature of the development of ML workflows. Within the ongoing ExtremeXP project (Horizon Europe), we propose an experiment-driven approach where systematic experimentation becomes central to ML workflow evolution. The framework created within the project supports transparent, reproducible, and adaptive experimentation through a formal metamodel and related domain-specific language. Key principles include traceable experiments for transparency, empowered decision-making for data scientists, and adaptive evolution through continuous feedback. In this paper, we present the framework from the model-based approach perspective. We discuss the lessons learned from the use of the metamodel-centric approach within the project-especially with use-case partners without prior modeling expertise.

Název v anglickém jazyce

A Model-Based Approach to Experiment-Driven Evolution of ML Workflows

Popis výsledku anglicky

Machine Learning (ML) has advanced significantly, yet the development of ML workflows still relies heavily on expert intuition, limiting standardization. MLOps integrates ML workflows for reliability, while AutoML automates tasks like hyperparameter tuning. However, these approaches often overlook the iterative and experimental nature of the development of ML workflows. Within the ongoing ExtremeXP project (Horizon Europe), we propose an experiment-driven approach where systematic experimentation becomes central to ML workflow evolution. The framework created within the project supports transparent, reproducible, and adaptive experimentation through a formal metamodel and related domain-specific language. Key principles include traceable experiments for transparency, empowered decision-making for data scientists, and adaptive evolution through continuous feedback. In this paper, we present the framework from the model-based approach perspective. We discuss the lessons learned from the use of the metamodel-centric approach within the project-especially with use-case partners without prior modeling expertise.

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Projekt

—

Návaznosti

S - Specificky vyzkum na vysokych skolach

Rok uplatnění

2025

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

International Conference on Model Driven Engineering and Software Development

ISBN

978-989-758-729-0

ISSN

2184-4348

e-ISSN

2184-4348

Počet stran výsledku

9

Strana od-do

354-362

Název nakladatele

scitepress

Místo vydání

Neuveden

Místo konání akce

Porto, Portugal

Datum konání akce

26. 2. 2025

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—