Human-interpretable model explainability on high-dimensional data
This addresses the challenge of explainability for complex models in domains like image classification, though it appears incremental by building on existing methods like Shapley values.
The authors tackled the problem of making model explainability computationally tractable and human-interpretable for high-dimensional data by introducing a framework that combines semantically meaningful latent representations with an adapted Shapley paradigm, demonstrating effectiveness on synthetic data and image-classification tasks.
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.