ATHENA: Advanced Techniques for High Dimensional Parameter Spaces to Enhance Numerical Analysis

TL;DR

ATHENA is an open source Python package that integrates advanced numerical techniques like Active Subspaces, Kernel-based Active Subspaces, and Nonlinear Level-set Learning to improve analysis in high-dimensional parameter spaces, aiding regression and sensitivity analysis.

Contribution

It introduces a comprehensive Python toolkit combining multiple advanced methods for high-dimensional parameter reduction and analysis.

Findings

Enables efficient regression and sensitivity analysis in high dimensions.

Provides open source tools with tutorials for practical application.

Facilitates tackling the curse of dimensionality in numerical simulations.

Abstract

ATHENA is an open source Python package for reduction in parameter space. It implements several advanced numerical analysis techniques such as Active Subspaces (AS), Kernel-based Active Subspaces (KAS), and Nonlinear Level-set Learning (NLL) method. It is intended as a tool for regression, sensitivity analysis, and in general to enhance existing numerical simulations' pipelines tackling the curse of dimensionality. Source code, documentation, and several tutorials are available on GitHub at https://github.com/mathLab/ATHENA under the MIT license.