This repository includes the codes for different projects on eXplainable Artificial Intelligence (XAI) by the main author.
Two Instances of Interpretable Neural Network for Universal Approximations (arxiv).
This project provides two bottom-up construction of interpretable universal approximations.
Fig. left A1-3,B1-3 show perfect accuracy for function approximations of Triangular Neural Network. Fig. right shows predictions based on interpolations of neighbouring values.
Enhancing the Confidence of Deep Learning Classifiers via Interpretable Saliency Maps (arxiv). The paper has also been published in the journal Neurocomputing!
This project attempts to improve predictive probability through the Augmentative Explanation process using existing heatmaps methods e.g. GradCAM and DeepLIFT.
This project provides synthetic dataset to measure the 'accuracy of heatmaps'.
Quantifying Explainability of Saliency Methods in Deep Neural Networks (IEEE TAI).
Fig. shows the synthetic dataset.
This project is for Generalization on the Enhancement of Layerwise Relevance Interpretability of Deep Neural Network. The project has been discontinued.
This subrepo contains the code for kaBEDONN. The project has been discontinued.
This subrepo contains the code for Evaluating Weakly Supervised Object Localization Methods Right? A Study on Heatmap-based XAI and Neural Backed Decision Tree. The project has been discontinued.