Towards Efficient and Explainable Automated Machine Learning Pipelines Design

Garouani, Moncef; Bouneffa, Mourad; Ahmad, Adeel

doi:10.54985/peeref.2208p4898652

Cited by 4 publications

(3 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As the KB is primarily utilized for storing and extracting data without the need for complex queries, and since each dataset, pipeline, and experiment may differ from one another, it can be structured in a NoSQL database format. This is similar to what was implemented in Garouani (2022), where the schema includes the entities that serve as the foundation for our KB.…”

Section: The Meta-knowledge Base Schemamentioning

confidence: 92%

“…Meta-learning based algorithm selection has been used in a variety of domains, such as automated machine learning , natural language processing Garouani andZaysa (2022), computer vision Mohammadi et al (2019), bioinformatics Arredondo and Ormazábal (2015) and many others. These applications are mainly focused on finding the best performing model for a given data set or task.…”

Section: Related Surveysmentioning

confidence: 99%

“…The knowledge base consists of several entities which are used to store and manage data during the inferring phase. The entities, based on a NoSQL database format used in previous works, Garouani (2022), are:…”

Section: The Meta-knowledge Base Schemamentioning

confidence: 99%

See 2 more Smart Citations

An experimental survey and Perspective View on Meta-Learning for Automated Algorithms Selection and Parametrization

Garouani,

Ahmad,

Bouneffa

2024

Preprint

Self Cite

View full text Add to dashboard Cite

Considerable progress has been made in the recent literature studies to tackle the Algorithms Selection and Parametrization (ASP) problem, which is diversified in multiple meta-learning setups. Yet there is a lack of surveys and comparative evaluations that critically analyze, summarize and assess the performance of existing methods. In this paper, we provide an overview of the state of the art in this continuously evolving field. The survey sheds light on the motivational reasons for pursuing classifiers selection through meta-learning. In this regard, Automated Machine Learning (AutoML) is usually treated as an ASP problem under the umbrella of the democratization of machine learning. Accordingly, AutoML makes machine learning techniques accessible to domain scientists who are interested in applying advanced analytics but lack the required expertise. It can ease the task of manually selecting ML algorithms and tuning related hyperparame-ters. We comprehensively discuss the different phases of classifiers selection based on a generic framework that is formed as an outcome of reviewing prior works. Subsequently, we propose a benchmark knowledge base of 4 millions previously learned models and present extensive comparative evaluations of the prominent methods for classifiers selection based on 08 classification algorithms and 400 benchmark datasets. The comparative study quantitatively assesses the performance of algorithms selection methods along while emphasizing the strengths and limitations of existing studies.

show abstract

Section: The Meta-knowledge Base Schemamentioning

confidence: 92%

Section: Related Surveysmentioning

confidence: 99%