SLISEMAP: Supervised dimensionality reduction through local explanations

Abstract

Existing methods for explaining black box learning models often focus on building local explanations of model behaviour for a particular data item. It is possible to create global explanations for all data items, but these explanations generally have low fidelity for complex black box models. We propose a new supervised manifold visualisation method, SLISEMAP, that simultaneously finds local explanations for all data items and builds a (typically) two-dimensional global visualisation of the black box model such that data items with similar local explanations are projected nearby. We provide a mathematical derivation of our problem and an open source implementation implemented using the GPU-optimised PyTorch library. We compare SLISEMAP to multiple popular dimensionality reduction methods and find that SLISEMAP is able to utilise labelled data to create embeddings with consistent local white box models. We also compare SLISEMAP to other model-agnostic local explanation methods and show that SLISEMAP provides comparable explanations and that the visualisations can give a broader understanding of black box regression and classification models.

arXiv comment26 pages, 10 figures, 3 tables. This revision replaces the ${\lambda }_z$ parameter with $z_{radius}$, which is more intuitive and stable. There are also many typographical and clarity improvements. The source code for our implementation of the algorithm (and the experiments) are available from GitHub at https://github.com/edahelsinki/slisemap . Machine Learning (2022)

References in corpus (13)

• PyTorch: An Imperative Style, High-Performance Deep Learning Library
• A Unified Approach to Interpreting Model Predictions
• Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localization
• European Union regulations on algorithmic decision-making and a "right to explanation"
• Searching for Exotic Particles in High-Energy Physics with Deep Learning
• All Models are Wrong, but Many are Useful: Learning a Variable's Importance by Studying an Entire Class of Prediction Models Simultaneously
• OpenML: networked science in machine learning
• Unmasking Clever Hans Predictors and Assessing What Machines Really Learn
• Interpretable Explanations of Black Boxes by Meaningful Perturbation
• How to Explain Individual Classification Decisions
• Linear Dimensionality Reduction: Survey, Insights, and Generalizations
• Dissecting stellar chemical abundance space with t-SNE
• SLISEMAP: Supervised dimensionality reduction through local explanations

Cited by in corpus (4)

• SLISEMAP: Supervised dimensionality reduction through local explanations
• Assessing feature scorer results on high-dimensional datasets with t-SNE
• $\chi$iplot: web-first visualisation platform for multidimensional data
• Using Slisemap to interpret physical data