Multi-Objective Evolutionary Design of Composite Data-Driven Models

Abstract

In this paper, a multi-objective approach for the design of composite data-driven mathematical models is proposed. It allows automating the identification of graph-based heterogeneous pipelines that consist of different blocks: machine learning models, data preprocessing blocks, etc. The implemented approach is based on a parameter-free genetic algorithm (GA) for model design called GPComp@Free. It is developed to be part of automated machine learning solutions and to increase the efficiency of the modeling pipeline automation. A set of experiments was conducted to verify the correctness and efficiency of the proposed approach and substantiate the selected solutions. The experimental results confirm that a multi-objective approach to the model design allows achieving better diversity and quality of obtained models. The implemented approach is available as a part of the open-source AutoML framework FEDOT.

Electrochemical Sensor to Detect Antibiotics in Milk Based on Machine Learning Algorithms

“…Classification and regression models were obtained in CatBoost [] and Fedot [] , frameworks. CatBoost is a version of the gradient boosting method based on learning each next ensemble model on a negative gradient, and the prediction models are typically simple decision trees.…”

Improvement of Computational Performance of Evolutionary AutoML in a Heterogeneous Environment

“…This architecture is implemented in the core of the opensource FEDOT framework. Different aspects of its implementation are already detailed in a series of papers: [19] describes the main schemes and the implementation of the evolutionary operators, [28] is devoted to the multi-objective modification of this approach, and [20] provides an extended description of the various aspects of the evolutionary design for composite modelling pipelines. The tuning strategy of the pipeline hyperparameters is based on Bayesian optimization.…”