Automated Tuning of Machine Learning Models in Real-World Applications

Authors: Ravi Kumar Perumallapalli

Country: USA

Full-text Research PDF File:   View   |   Download

Abstract: The growing intricacy of machine learning (ML) models and their extensive use in practical applica-tions demand effective and expandable methods for automating model adjustment. This study inves-tigates the most recent approaches and technologies for automatic machine learning (AutoML), with a particular emphasis on methods that optimize model topologies and hyperparameters in real-world scenarios. This work analyzes existing solutions and their efficacy across several applications, leverag-ing recent improvements in AutoML frameworks, such as OpenML benchmarking suites and state-of-the-art research on automation issues. Various data distributions, noisy settings, and the requirement for model adaptation make it difficult to integrate automated tuning in real-world systems. By means of a comparative examination of AutoML tools we exhibit how these techniques simplify the process of developing and implementing models, providing significant perspectives on augmenting productivity in both scientific and industrial fields.

Keywords: AutoML, machine learning, model optimization, hyperparameter tuning, real-world applications, OpenML, benchmarking, automation, data distribution, noisy environments, model adaptation, prod-uctivity enhancement, industrial applications.

Paper Id: 231515

Published On: 2019-01-03

Published In: Volume 7, Issue 1, January-February 2019

Cite This: Automated Tuning of Machine Learning Models in Real-World Applications - Ravi Kumar Perumallapalli - IJIRMPS Volume 7, Issue 1, January-February 2019.

Share this