DISSECT: Disentangled Simultaneous Explanations via Concept Traversals

by   Asma Ghandeharioun, et al.

Explaining deep learning model inferences is a promising venue for scientific understanding, improving safety, uncovering hidden biases, evaluating fairness, and beyond, as argued by many scholars. One of the principal benefits of counterfactual explanations is allowing users to explore "what-if" scenarios through what does not and cannot exist in the data, a quality that many other forms of explanation such as heatmaps and influence functions are inherently incapable of doing. However, most previous work on generative explainability cannot disentangle important concepts effectively, produces unrealistic examples, or fails to retain relevant information. We propose a novel approach, DISSECT, that jointly trains a generator, a discriminator, and a concept disentangler to overcome such challenges using little supervision. DISSECT generates Concept Traversals (CTs), defined as a sequence of generated examples with increasing degrees of concepts that influence a classifier's decision. By training a generative model from a classifier's signal, DISSECT offers a way to discover a classifier's inherent "notion" of distinct concepts automatically rather than rely on user-predefined concepts. We show that DISSECT produces CTs that (1) disentangle several concepts, (2) are influential to a classifier's decision and are coupled to its reasoning due to joint training (3), are realistic, (4) preserve relevant information, and (5) are stable across similar inputs. We validate DISSECT on several challenging synthetic and realistic datasets where previous methods fall short of satisfying desirable criteria for interpretability and show that it performs consistently well and better than existing methods. Finally, we present experiments showing applications of DISSECT for detecting potential biases of a classifier and identifying spurious artifacts that impact predictions.


page 2

page 9

page 19

page 20


ConceptDistil: Model-Agnostic Distillation of Concept Explanations

Concept-based explanations aims to fill the model interpretability gap f...

Unsupervised Causal Binary Concepts Discovery with VAE for Black-box Model Explanation

We aim to explain a black-box classifier with the form: `data X is class...

STEEX: Steering Counterfactual Explanations with Semantics

As deep learning models are increasingly used in safety-critical applica...

Using Causal Analysis for Conceptual Deep Learning Explanation

Model explainability is essential for the creation of trustworthy Machin...

Meaningfully Explaining a Model's Mistakes

Understanding and explaining the mistakes made by trained models is crit...

Enriching Visual with Verbal Explanations for Relational Concepts – Combining LIME with Aleph

With the increasing number of deep learning applications, there is a gro...

Explaining Deep Graph Networks with Molecular Counterfactuals

We present a novel approach to tackle explainability of deep graph netwo...

Code Repositories


DISSECT: Disentangled Simultaneous Explanations via Concept Traversals

view repo