Model-Based Counterfactual Synthesizer for Interpretation

Abstract

Counterfactuals, serving as one of the emerging type of model interpretations, have recently received attention from both researchers and practitioners. Counterfactual explanations formalize the exploration of ``what-if'' scenarios, and are an instance of example-based reasoning using a set of hypothetical data samples. Counterfactuals essentially show how the model decision alters with input perturbations. Existing methods for generating counterfactuals are mainly algorithm-based, which are time-inefficient and assume the same counterfactual universe for different queries. To address these limitations, we propose a Model-based Counterfactual Synthesizer (MCS) framework for interpreting machine learning models. We first analyze the model-based counterfactual process and construct a base synthesizer using a conditional generative adversarial net (CGAN). To better approximate the counterfactual universe for those rare queries, we novelly employ the umbrella sampling technique to conduct the MCS framework training. Besides, we also enhance the MCS framework by incorporating the causal dependence among attributes with model inductive bias, and validate its design correctness from the causality identification perspective. Experimental results on several datasets demonstrate the effectiveness as well as efficiency of our proposed MCS framework, and verify the advantages compared with other alternatives.

11 pages in total; To appear in KDD'21;

References in corpus (16)

• A Survey on Deep Learning in Medical Image Analysis
• Conditional Generative Adversarial Nets
• A Unified Approach to Interpreting Model Predictions
• Categorical Reparameterization with Gumbel-Softmax
• Axiomatic Attribution for Deep Networks
• Understanding Black-box Predictions via Influence Functions
• Explaining Machine Learning Classifiers through Diverse Counterfactual Explanations
• This Looks Like That: Deep Learning for Interpretable Image Recognition
• Data Synthesis based on Generative Adversarial Networks
• Learning Model-Agnostic Counterfactual Explanations for Tabular Data
• Preserving Causal Constraints in Counterfactual Explanations for Machine Learning Classifiers
• Explaining Classifiers with Causal Concept Effect (CaCE)
• Towards Automatic Concept-based Explanations
• Measurable Counterfactual Local Explanations for Any Classifier
• A survey of algorithmic recourse: definitions, formulations, solutions, and prospects
• xGEMs: Generating Examplars to Explain Black-Box Models

Cited by in corpus (4)

• Achieving Diversity in Counterfactual Explanations: a Review and Discussion
• RoCourseNet: Distributionally Robust Training of a Prediction Aware Recourse Model
• For Better or Worse: The Impact of Counterfactual Explanations' Directionality on User Behavior in xAI
• MCCE: Monte Carlo sampling of realistic counterfactual explanations