Causal Proxy Models For Concept-Based Model Explanations

Zhengxuan Wu, Karel D'Oosterlinck, Atticus Geiger, Amir Zur, Christopher Potts

Keywords: Explainability, Causality, Concept-Based Explanations, Causal Explanations

TL;DR: We introduce Causal Proxy Models (CPMs), a novel concept-based explanation method. Our method used counterfactual training data to achieve state-of-the-art explanation performance on the CEBaB benchmark.

Abstract: Explainability methods for NLP systems encounter a version of the fundamental problem of causal inference: for a given ground-truth input text, we never truly observe the counterfactual texts necessary for isolating the causal effects of model representations on outputs. In response, many explainability methods make no use of counterfactual texts, assuming they will be unavailable. In this paper, we show that robust causal explainability methods can be created using approximate counterfactuals, which can be written by humans to approximate a specific counterfactual or simply sampled using metadata-guided heuristics. The core of our proposal is the Causal Proxy Model (CPM). A CPM explains a black-box model $\mathcal{N}$ because it is trained to have the same \emph{actual} input/output behavior as $\mathcal{N}$ while creating neural representations that can be intervened upon to simulate the \emph{counterfactual} input/output behavior of $\mathcal{N}$. Furthermore, we show that the best CPM for $\mathcal{N}$ performs comparably to $\mathcal{N}$ in making factual predictions, which means that the CPM can simply replace $\mathcal{N}$, leading to more explainable deployed models.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics

Submission Guidelines: Yes

Please Choose The Closest Area That Your Submission Falls Into: Social Aspects of Machine Learning (eg, AI safety, fairness, privacy, interpretability, human-AI interaction, ethics)

Community Implementations: [ 1 code implementation](https://www.catalyzex.com/paper/causal-proxy-models-for-concept-based-model/code)