MCCE: Missingness-aware Causal Concept Explainer

TL;DR

MCCE is a novel framework that accurately estimates causal effects of concepts in machine learning models even when some concepts are missing, improving interpretability despite incomplete data.

Contribution

It introduces a missingness-aware causal explanation method that accounts for unobserved concepts, addressing a key limitation of existing approaches.

Findings

MCCE outperforms state-of-the-art methods in causal effect estimation.

It effectively handles incomplete concept annotations in real-world data.

Provides both local and global explanations of model behavior.

Abstract

Causal concept effect estimation is gaining increasing interest in the field of interpretable machine learning. This general approach explains the behaviors of machine learning models by estimating the causal effect of human-understandable concepts, which represent high-level knowledge more comprehensibly than raw inputs like tokens. However, existing causal concept effect explanation methods assume complete observation of all concepts involved within the dataset, which can fail in practice due to incomplete annotations or missing concept data. We theoretically demonstrate that unobserved concepts can bias the estimation of the causal effects of observed concepts. To address this limitation, we introduce the Missingness-aware Causal Concept Explainer (MCCE), a novel framework specifically designed to estimate causal concept effects when not all concepts are observable. Our framework learns to account for residual bias resulting from missing concepts and utilizes a linear predictor to model the relationships between these concepts and the outputs of black-box machine learning models. It can offer explanations on both local and global levels. We conduct validations using a real-world dataset, demonstrating that MCCE achieves promising performance compared to state-of-the-art explanation methods in causal concept effect estimation.