Using Slisemap to interpret physical data

Abstract

Manifold visualisation techniques are commonly used to visualise high-dimensional datasets in physical sciences. In this paper we apply a recently introduced manifold visualisation method, called Slise, on datasets from physics and chemistry. Slisemap combines manifold visualisation with explainable artificial intelligence. Explainable artificial intelligence is used to investigate the decision processes of black box machine learning models and complex simulators. With Slisemap we find an embedding such that data items with similar local explanations are grouped together. Hence, Slisemap gives us an overview of the different behaviours of a black box model. This makes Slisemap into a supervised manifold visualisation method, where the patterns in the embedding reflect a target property. In this paper we show how Slisemap can be used and evaluated on physical data and that Slisemap is helpful in finding meaningful information on classification and regression models trained on these datasets.

arXiv comment17 pages, 5 + 1 figures, 1 table. The datasets and source code used in the paper are available at https://www.edahelsinki.fi/papers/slisemap_phys

References in corpus (12)

• A Unified Approach to Interpreting Model Predictions
• UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction
• Machine learning and the physical sciences
• Fast and Accurate Modeling of Molecular Atomization Energies with Machine Learning
• Linear Dimensionality Reduction: Survey, Insights, and Generalizations
• A survey of dimensionality reduction techniques
• Dissecting stellar chemical abundance space with t-SNE
• To trust or not to trust an explanation: using LEAF to evaluate local linear XAI methods
• Chemical diversity in molecular orbital energy predictions with kernel ridge regression
• Unrestricted Permutation forces Extrapolation: Variable Importance Requires at least One More Model, or There Is No Free Variable Importance
• SLISEMAP: Supervised dimensionality reduction through local explanations
• $\chi$iplot: web-first visualisation platform for multidimensional data