This is an academic repository for CITS4404 - Artificial Intelligence and Adaptive Systems
Increased computing power allows development of increasingly accurate machine learning models. This increase in accuracy is generally accompanied by an increase in model complexity. We often have a blackbox model', a model which receives inputs and makes outputs, or predictions, with little visibility possible into the mechanics of the model, or reasons a prediction was made. Explainable Artificial Intelligence' or XAI' is an active research area, seeking to develop methods to explain the reasons for model output to users. One method of explanation is to use Counterfactual Explanations'. Herein we examine some currently available python modules that cater for Counterfactual Explanations, and select some for further analysis and implementation.
Link to Github repository: https://github.com/wiherreira/XAI-CounterfactualExplanations