Using slisemap to interpret physical data

Journal article

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.Comment: 17 pages, 5 + 1 figures, 1 table. The datasets and source code used in the paper are available at https://www.edahelsinki.fi/papers/slisemap_phy

Authors: Seppäläinen, L; Björklund, A; Besel, V; Puolamäki, K

• Publication status: Published
• Peer review status: Peer reviewed
• Publisher: Public Library of Science
• Journal: PLoS ONE
• Volume: 19
• Issue: 1
• Pages: e0297714-e0297714
• Publication date: 2024-01-25
• DOI: 10.1371/journal.pone.0297714
• EISSN: 1932-6203
• ISSN: 1932-6203
• Language: English
• Keywords: Visualization; Artificial intelligence; Nonlinear dimensionality reduction; Manifold (fluid mechanics); Information retrieval; Black box; Data science; Property (philosophy); Machine learning; Data visualization; Dimensionality reduction; Embedding; Computer science; Data mining