Auxiliary Task-Based Multi-Task Evolutionary Algorithm for High-Dimensional Feature Selection in Classification

With the development of machine learning, feature selection has become a crucial step to improve model performance and interpretability. However, traditional feature selection methods often face issues such as the curse of dimensionality and low computational efficiency. To address these issues, evolutionary computation (EC) has received widespread attention due to its efficient global search capability. Nevertheless, the diverse design of EC methods results in varying adaptability when dealing with different types of data, which typically leads to insufficient information utilization and difficulty in achieving effective information sharing. This paper proposes an evolutionary multi-task evolutionary algorithm framework. This framework leverages diverse knowledge from auxiliary tasks to enhance the global exploration capability of the main task. By integrating information sharing among different feature selection tasks, MBBPSO enhances the learning ability and efficiency of the algorithm. Experimental results show that compared with other feature selection algorithms on 8 datasets, the proposed MBBPSO method demonstrates strong competitiveness in both classification accuracy and runtime.

• Publication year

  2025

• Venue

  2025 7th International Conference on Data-driven Optimization of Complex Systems (DOCS)

• Publication date

  2025-08-19

• Fields of study

  Not labeled

• Identifiers
  DOI 10.1109/DOCS67533.2025.11200923
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Particle swarm optimization algorithm based on comprehensive scoring framework for high-dimensional feature selection

  2025cited by this paper
• Two-Stage Hybrid Feature Selection: Integrating ACO Algorithms With a Statistical Ensemble Technique for EV Demand Prediction

  2025cited by this paper
• TS-RePSO: A Three-Stage Feature Selection Method Combing ReliefF and PSO in Bioinformatics

  2025cited by this paper
• A new filter feature selection algorithm for classification task by ensembling pearson correlation coefficient and mutual information

  2024cited by this paper
• An enhanced particle swarm optimization with position update for optimal feature selection

  2024cited by this paper
• A memory interaction quadratic interpolation whale optimization algorithm based on reverse information correction for high-dimensional feature selection

  2024cited by this paper
• Search space division method for wrapper feature selection on high-dimensional data classification

  2024cited by this paper
• MEL: Efficient Multi-Task Evolutionary Learning for High-Dimensional Feature Selection

  2024cited by this paper
• Analysis and comparison of feature selection methods towards performance and stability

  2024cited by this paper
• An evolutionary multiobjective method based on dominance and decomposition for feature selection in classification

  2024cited by this paper
• Towards Multi-Objective High-Dimensional Feature Selection via Evolutionary Multitasking

  2024cited by this paper
• A bidirectional dynamic grouping multi-objective evolutionary algorithm for feature selection on high-dimensional classification

  2023cited by this paper
• Explicit and size-adaptive PSO-based feature selection for classification

  2023cited by this paper
• SFE: A Simple, Fast, and Efficient Feature Selection Algorithm for High-Dimensional Data

  2023cited by this paper
• A novel autoencoder based feature independent GA optimised XGBoost classifier for IoMT malware detection

  2023cited by this paper
• Multitask Particle Swarm Optimization With Dynamic Transformation

  2023cited by this paper
• An Evolutionary Multitasking Algorithm With Multiple Filtering for High-Dimensional Feature Selection

  2022cited by this paper
• Multi-surrogate assisted multi-objective evolutionary algorithms for feature selection in regression and classification problems with time series data

  2022cited by this paper
• Evolutionary Multitasking for Feature Selection in High-Dimensional Classification via Particle Swarm Optimization

  2021cited by this paper
• An Evolutionary Multitasking-Based Feature Selection Method for High-Dimensional Classification

  2020cited by this paper
• Binary differential evolution with self-learning for multi-objective feature selection

  2020cited by this paper
• Cost-sensitive feature selection using two-archive multi-objective artificial bee colony algorithm

  2019cited by this paper
• Feature selection algorithm based on bare bones particle swarm optimization

  2015cited by this paper
• Bare bones particle swarms

  2003cited by this paper

• No citing papers are available for this paper.