Integrated Gradient attribution for Gaussian Processes with non-Gaussian likelihoods
This work addresses the need for transparent decision-making in probabilistic machine learning, particularly for Gaussian Processes used in complex data scenarios, though it is incremental as it extends existing gradient-based explainability to a specific model type.
The paper tackles the challenge of applying gradient-based explainability methods like Integrated Gradients to Gaussian Process models with non-Gaussian likelihoods, such as in classification, by proposing analytical and approximate solutions to handle the non-trivial partial derivatives involved.
Gaussian Process (GP) models are a powerful tool in probabilistic machine learning with a solid theoretical foundation. Thanks to current advances, modeling complex data with GPs is becoming increasingly feasible, which makes them an interesting alternative to deep learning and related approaches. As the latter are getting more and more influential on society, the need for making a model's decision making process transparent and explainable is now a major focus of research. A major direction in interpretable machine learning is the use of gradient-based approaches, such as Integrated Gradients, to quantify feature attribution, locally for a given datapoint of interest. Since GPs and the behavior of their partial derivatives are well studied and straightforward to derive, studying gradient-based explainability for GPs is a promising direction of research. Unfortunately, partial derivatives for GPs become less trivial to handle when dealing with non-Gaussian target data as in classification or more sophisticated regression problems. This paper therefore proposes an approach for applying Integrated Gradient-based explainability to non-Gaussian GP models, offering both analytical and approximate solutions. This extends gradient-based explainability to probabilistic models with complex likelihoods to extend their practical applicability.