DeepAI AI Chat
Log In Sign Up

Human-interpretable model explainability on high-dimensional data

by   Damien de Mijolla, et al.

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.


page 5

page 7

page 8

page 13

page 14

page 15

page 16


Shapley-based explainability on the data manifold

Explainability in machine learning is crucial for iterative model develo...

Explaining Predictions by Approximating the Local Decision Boundary

Constructing accurate model-agnostic explanations for opaque machine lea...

Algorithm-Agnostic Explainability for Unsupervised Clustering

Supervised machine learning explainability has greatly expanded in recen...

Model-Agnostic Explainability for Visual Search

What makes two images similar? We propose new approaches to generate mod...

Towards Ground Truth Explainability on Tabular Data

In data science, there is a long history of using synthetic data for met...

On quantitative aspects of model interpretability

Despite the growing body of work in interpretable machine learning, it r...