Pixel-level feature attributions are an important tool in eXplainable AI for Computer Vision (XCV), providing visual insights into how image features influence model predictions. The Owen formula for hierarchical Shapley values has been widely used to interpret machine learning (ML) models and their learned representations. However, existing hierarchical Shapley approaches do not exploit the multiscale structure of image data, leading to slow convergence and weak alignment with the actual morphological features. Moreover, no prior Shapley method has leveraged data-aware hierarchies for Computer Vision tasks, leaving a gap in model interpretability of structured visual data. To address this, this paper introduces ShapBPT, a novel data-aware XCV method based on the hierarchical Shapley formula. ShapBPT assigns Shapley coefficients to a multiscale hierarchical structure tailored for images, the Binary Partition Tree (BPT). By using this data-aware hierarchical partitioning, ShapBPT ensures that feature attributions align with intrinsic image morphology, effectively prioritizing relevant regions while reducing computational overhead. This advancement connects hierarchical Shapley methods with image data, providing a more efficient and semantically meaningful approach to visual interpretability. Experimental results confirm ShapBPT’s effectiveness, demonstrating superior alignment with image structures and improved efficiency over existing XCV methods, and a 20-subject user study confirming that ShapBPT explanations are preferred by humans.

Data augmentation represents an opportunity for Artificial Intelligence (AI) applications, as it aims at creating new synthetic data based on an existing baseline. In this paper, we present a new evaluation framework for Generative Adversarial Networks (GANs), a data augmentation technique, in multivariate data classification contexts. The goal is not limited to assessing the performance variations obtained through GANs, but also to inspect results with explainable AI (XAI) tools, understanding how GANs work and, finally, exploiting them to discover new knowledge. To this aim, we adopt the Logic Learning Machine (LLM) for performance assessment and rule extraction, and introduce a new measure of rule similarity to compare different artificial datasets. We apply the methodology on two case studies , activity recognition and physical fatigue detection, confirming that GANs can help in overcoming limitations of original datasets and lead to new discoveries.

We investigate the use of a stratified sampling approach for LIME Image, a popular model-agnostic explainable AI method for computer vision tasks, in order to reduce the artifacts generated by typical Monte Carlo sampling. Such artifacts are due to the undersampling of the dependent variable in the synthetic neighborhood around the image being explained, which may result in inadequate explanations due to the impossibility of fitting a linear regressor on the sampled data. We then highlight a connection with the Shapley theory, where similar arguments about undersampling and sample relevance were suggested in the past. We derive all the formulas and adjustment factors required for an unbiased stratified sampling estimator. Experiments show the efficacy of the proposed approach.

Data augmentation is a widespread innovative technique in Artificial Intelligence: it aims at creating new synthetic data given an existing real baseline, thus allowing to overcome the issues arising from the lack of labelled data for proper training of classification algorithms. Our paper focuses on how a common data augmentation methodology, the Generative Adversarial Networks (GANs), which is widespread for images and timeseries data, can be also applied to generate multivariate data. We propose a novel scheme for GANs evaluation, based on the performance of an explainable AI (XAI) algorithm and an innovative definition of rule similarity. In particular, we will consider an application dealing with the augmentation of Inertial Movement Units (IMU) data for physical fatigue monitoring in two age subgroups (under and over 40 years old) of the original data. We will show how our innovative rule similarity metric can drive the selection of the best fake dataset among a set of different candidates, corresponding to different GAN training runs.

BACKGROUND: Logic Learning Machine (LLM) is an innovative method of supervised analysis capable of constructing models based on simple and intelligible rules. In this investigation the performance of LLM in classifying patients with cancer was evaluated using a set of eight publicly available gene expression databases for cancer diagnosis. LLM accuracy was assessed by summary ROC curve (sROC) analysis and estimated by the area under an sROC curve (sAUC). Its performance was compared in cross validation with that of standard supervised methods, namely: decision tree, artificial neural network, support vector machine (SVM) and k-nearest neighbor classifier. RESULTS: LLM showed an excellent accuracy (sAUC = 0.99, 95%CI: 0.98-1.0) and outperformed any other method except SVM. CONCLUSIONS: LLM is a new powerful tool for the analysis of gene expression data for cancer diagnosis. Simple rules generated by LLM could contribute to a better understanding of cancer biology, potentially addressing therapeutic approaches.

In this paper, we address the problem of localizing a camera-equipped Micro Aerial Vehicle (MAV) flying in urban streets at low altitudes. An appearance-based global positioning system to localize MAVs with respect to the surrounding buildings is introduced. We rely on an air-ground image matching algorithm to search the airborne image of the MAV within a ground-level Street View image database and to detect image matching points. Based on the image matching points, we infer the global position of the MAV by back-projecting the corresponding image points onto a cadastral 3D city model. Furthermore, we describe an algorithm to track the position of the flying vehicle over several frames and to correct the accumulated drift of the visual odometry, whenever a good match is detected between the airborne MAV and the street-level images. The proposed approach is tested on a dataset captured with a small quadroctopter flying in the streets of Zurich.