An Explainable Gaussian Process Auto-encoder for Tabular Data
This work addresses the need for explainable AI in high-stakes domains, offering an incremental improvement over existing auto-encoder-based methods.
The authors tackled the problem of generating counterfactual explanations for black-box models by proposing a Gaussian process auto-encoder for tabular data, which produced diversified and in-distribution samples in experiments on large-scale datasets.
Explainable machine learning has attracted much interest in the community where the stakes are high. Counterfactual explanations methods have become an important tool in explaining a black-box model. The recent advances have leveraged the power of generative models such as an autoencoder. In this paper, we propose a novel method using a Gaussian process to construct the auto-encoder architecture for generating counterfactual samples. The resulting model requires fewer learnable parameters and thus is less prone to overfitting. We also introduce a novel density estimator that allows for searching for in-distribution samples. Furthermore, we introduce an algorithm for selecting the optimal regularization rate on density estimator while searching for counterfactuals. We experiment with our method in several large-scale tabular datasets and compare with other auto-encoder-based methods. The results show that our method is capable of generating diversified and in-distribution counterfactual samples.