Is Disentanglement all you need? Comparing Concept-based Disentanglement Approaches

by   Dmitry Kazhdan, et al.

Concept-based explanations have emerged as a popular way of extracting human-interpretable representations from deep discriminative models. At the same time, the disentanglement learning literature has focused on extracting similar representations in an unsupervised or weakly-supervised way, using deep generative models. Despite the overlapping goals and potential synergies, to our knowledge, there has not yet been a systematic comparison of the limitations and trade-offs between concept-based explanations and disentanglement approaches. In this paper, we give an overview of these fields, comparing and contrasting their properties and behaviours on a diverse set of tasks, and highlighting their potential strengths and limitations. In particular, we demonstrate that state-of-the-art approaches from both classes can be data inefficient, sensitive to the specific nature of the classification/regression task, or sensitive to the employed concept representation.


Deep Generative Models for Physiological Signals: A Systematic Literature Review

In this paper, we present a systematic literature review on deep generat...

Concept-Oriented Deep Learning: Generative Concept Representations

Generative concept representations have three major advantages over disc...

Concept-Based Techniques for "Musicologist-friendly" Explanations in a Deep Music Classifier

Current approaches for explaining deep learning systems applied to music...

Automated patent extraction powers generative modeling in focused chemical spaces

Deep generative models have emerged as an exciting avenue for inverse mo...

Investigating Bias in Image Classification using Model Explanations

We evaluated whether model explanations could efficiently detect bias in...

Unsupervised Interpretable Basis Extraction for Concept-Based Visual Explanations

An important line of research attempts to explain CNN image classifier p...

Unsupervised Learning of Structured Representations via Closed-Loop Transcription

This paper proposes an unsupervised method for learning a unified repres...

Please sign up or login with your details

Forgot password? Click here to reset