EDCA -- An Evolutionary Data-Centric AutoML Framework for Efficient Pipelines

2025-03-06

AI Summary

Traditional AutoML frameworks largely overlook the impact of data quality on model performance, while manual data cleaning and dimensionality reduction remain inefficient and labor-intensive. Method: This paper introduces a data-centric AutoML paradigm that integrates data cleaning and dimensionality reduction into an end-to-end optimization pipeline. Data transformation operations are encoded as evolvable graph-structured nodes, enabling joint optimization of preprocessing strategies and model hyperparameters. Multi-objective optimization is performed using NSGA-II to simultaneously maximize predictive accuracy and minimize training data volume and computational cost—aligning with Green AutoML principles. Contribution/Results: Evaluated on the AMLB classification benchmark, our approach achieves accuracy comparable to state-of-the-art systems (FLAML, TPOT), yet requires significantly less training data. This demonstrates that synergistic co-optimization of high-quality data subsets and algorithmic components can attain equivalent performance with reduced resource overhead.

Technology Category

• Machine Learning: Auto ML and Hyperparameter Tuning
• Search and Optimization: Learning to Search
• Natural Language Processing: Learning & Optimization for NLP

Application Category

• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Web Mining and Content Analysis: Web data integration and cleaning
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs

Abstract

Automated Machine Learning (AutoML) gained popularity due to the increased demand for Machine Learning (ML) specialists, allowing them to apply ML techniques effortlessly and quickly. AutoML implementations use optimisation methods to identify the most effective ML solution for a given dataset, aiming to improve one or more predefined metrics. However, most implementations focus on model selection and hyperparameter tuning. Despite being an important factor in obtaining high-performance ML systems, data quality is usually an overlooked part of AutoML and continues to be a manual and time-consuming task. This work presents EDCA, an Evolutionary Data Centric AutoML framework. In addition to the traditional tasks such as selecting the best models and hyperparameters, EDCA enhances the given data by optimising data processing tasks such as data reduction and cleaning according to the problems' needs. All these steps create an ML pipeline that is optimised by an evolutionary algorithm. To assess its effectiveness, EDCA was compared to FLAML and TPOT, two frameworks at the top of the AutoML benchmarks. The frameworks were evaluated in the same conditions using datasets from AMLB classification benchmarks. EDCA achieved statistically similar results in performance to FLAML and TPOT but used significantly less data to train the final solutions. Moreover, EDCA experimental results reveal that a good performance can be achieved using less data and efficient ML algorithm aspects that align with Green AutoML guidelines

Problem

• Automates data processing and model optimization in AutoML.
• Focuses on data quality improvement for efficient ML pipelines.
• Reduces data usage while maintaining high performance in ML.

Innovation

• Evolutionary algorithm optimizes ML pipelines
• Enhances data quality via reduction and cleaning
• Achieves high performance with less data usage