Structure Your Data: Towards Semantic Graph Counterfactuals

arXiv:2403.06514v1 Announce Type: new
Abstract: Counterfactual explanations (CEs) based on concepts are explanations that consider alternative scenarios to understand which high-level semantic features contributed to particular model predictions. In this work, we propose CEs based on the semantic graphs accompanying input data to achieve more descriptive, accurate, and human-aligned explanations. Building upon state-of-the-art (SoTA) conceptual attempts, we adopt a model-agnostic edit-based approach and introduce leveraging GNNs for efficient Graph Edit Distance (GED) computation. With a focus on the visual domain, …

abstract art arxiv building ces concepts contributed counterfactual cs.ai cs.cv data features graph graphs human predictions semantic state type work