Human-interpretable model explainability on high-dimensional data

Abstract

The importance of explainability in machine learning continues to grow, as both neural-network architectures and the data they model become increasingly complex. Unique challenges arise when a model's input features become high dimensional: on one hand, principled model-agnostic approaches to explainability become too computationally expensive; on the other, more efficient explainability algorithms lack natural interpretations for general users. In this work, we introduce a framework for human-interpretable explainability on high-dimensional data, consisting of two modules. First, we apply a semantically meaningful latent representation, both to reduce the raw dimensionality of the data, and to ensure its human interpretability. These latent features can be learnt, e.g. explicitly as disentangled representations or implicitly through image-to-image translation, or they can be based on any computable quantities the user chooses. Second, we adapt the Shapley paradigm for model-agnostic explainability to operate on these latent features. This leads to interpretable model explanations that are both theoretically controlled and computationally tractable. We benchmark our approach on synthetic data and demonstrate its effectiveness on several image-classification tasks.

8 pages, 6 figures, 1 appendix

References in corpus (4)

• Visualizing Deep Neural Network Decisions: Prediction Difference Analysis
• Fader Networks: Manipulating Images by Sliding Attributes
• Towards Frequency-Based Explanation for Robust CNN
• Shapley explainability on the data manifold

Cited by in corpus (3)

• Does Your Model Think Like an Engineer? Explainable AI for Bearing Fault Detection with Deep Learning
• The Thousand Faces of Explainable AI Along the Machine Learning Life Cycle: Industrial Reality and Current State of Research
• Deep Learning Based Decision Support for Medicine -- A Case Study on Skin Cancer Diagnosis