Explaining machine learning classifiers through diverse counterfactual explanations

Abstract

Post-hoc explanations of machine learning models are crucial for people to understand and act on algorithmic predictions. An intriguing class of explanations is through counterfactuals, hypothetical examples that show people how to obtain a different prediction. We posit that effective counterfactual explanations should satisfy two properties: feasibility of the counterfactual actions given user context and constraints, and diversity among the counterfactuals presented. To this end, we propose a framework for generating and evaluating a diverse set of counterfactual explanations based on average distance and determinantal point processes. To evaluate the actionability of counterfactuals, we provide metrics that enable comparison of counterfactual-based methods to other local explanation methods. We further address necessary tradeoffs and point to causal implications in optimizing for counterfactuals. Our experiments on three real-world datasets show that our framework can generate a set of counterfactuals that are diverse and well approximate local decision boundaries.

Multi-Objective Counterfactual Explanations

“…For the first one, we compared MOC -once with and once without our proposed strategies for initialization and mutation -with 'DiCE' by Mothilal et al [24], 'Recourse' by Ustun et al [33] and 'Tweaking' by Tolomei et al [32]. We chose DiCE, Recourse and Tweaking because they are implemented in general open source code libraries.…”

“…A model-specific approach was proposed by Wachter et al [35], who also introduced and formalized the concept of counterfactuals in predictive modeling. Like many model-specific methods [15,20,24,28,33] their approach is limited to differentiable models. The approach of Tolomei et al [32] generates explanations for tree-based ensemble binary classifiers.…”

Introduction to digital pathology and computer-aided pathology

“…In applications to pathologic image analysis, an attention mechanism was used to visualize epithelial cell areas [57]. Arbitrary generated counterfactual examples can be used to describe how input change affects to model output [58]. Case-based reasoning [59,60] can be combined with the interpretation target model to demonstrate consistency between model output and reasoning output of the same input [61].…”

ViCE

“…Comparing the data values to the density distributions depicted by the shaded area can help identify anomalies and derive hypotheses on why and how the model produces a given prediction. It is also worth noticing that the visualization interface can accommodate any other counterfactual generation methods [9,16,21,22].…”