Research Papers

xAI-Drop: Don't Use What You Cannot Explain

Vincenzo Marco De Luca, Antonio Longa, Pietro Liò +1

Graph Neural Networks (GNNs) have emerged as the predominant paradigm for learning from graph-structured data, offering a wide range of applications from social network analysis to bioinformatics. Despite their versatility, GNNs face challenges such as lack of generalization and poor interpretability, which hinder their wider adoption and reliability in critical applications. Dropping has emerged as an effective paradigm for improving the generalization capabilities of GNNs. However, existing approaches often rely on random or heuristic-based selection criteria, lacking a principled method to identify and exclude nodes that contribute to noise and over-complexity in the model. In this work, we argue that explainability should be a key indicator of a model's quality throughout its training phase. To this end, we introduce xAI-Drop, a novel topological-level dropping regularizer that leverages explainability to pinpoint noisy network elements to be excluded from the GNN propagation mechanism. An empirical evaluation on diverse real-world datasets demonstrates that our method outperforms current state-of-the-art dropping approaches in accuracy, and improves explanation quality.

Different prediction models might perform equally well (Rashomon set) in the same task, but offer conflicting interpretations and conclusions about the data. The Rashomon effect in the context of Explainable AI (XAI) has been recognized as a critical factor. Although the Rashomon set has been introduced and studied in various contexts, its practical application is at its infancy stage and lacks adequate guidance and evaluation. We study the problem of the Rashomon set sampling from a practical viewpoint and identify two fundamental axioms - generalizability and implementation sparsity that exploring methods ought to satisfy in practical usage. These two axioms are not satisfied by most known attribution methods, which we consider to be a fundamental weakness. We use the norms to guide the design of an $ε$-subgradient-based sampling method. We apply this method to a fundamental mathematical problem as a proof of concept and to a set of practical datasets to demonstrate its ability compared with existing sampling methods.

Explainable Artificial Intelligence (XAI) aims to improve the transparency of autonomous decision-making through explanations. Recent literature has emphasised users' need for holistic "multi-shot" explanations and personalised engagement with XAI systems. We refer to this user-centred interaction as an XAI Experience. Despite advances in creating XAI experiences, evaluating them in a user-centred manner has remained challenging. In response, we developed the XAI Experience Quality (XEQ) Scale. XEQ quantifies the quality of experiences across four dimensions: learning, utility, fulfilment and engagement. These contributions extend the state-of-the-art of XAI evaluation, moving beyond the one-dimensional metrics frequently developed to assess single-shot explanations. This paper presents the XEQ scale development and validation process, including content validation with XAI experts, and discriminant and construct validation through a large-scale pilot study. Our pilot study results offer strong evidence that establishes the XEQ Scale as a comprehensive framework for evaluating user-centred XAI experiences.

Machine learned potentials are becoming a popular tool to define an effective energy model for complex systems, either incorporating electronic structure effects at the atomistic resolution, or effectively renormalizing part of the atomistic degrees of freedom at a coarse-grained resolution. One of the main criticisms to machine learned potentials is that the energy inferred by the network is not as interpretable as in more traditional approaches where a simpler functional form is used. Here we address this problem by extending tools recently proposed in the nascent field of Explainable Artificial Intelligence (XAI) to coarse-grained potentials based on graph neural networks (GNN). We demonstrate the approach on three different coarse-grained systems including two fluids (methane and water) and the protein NTL9. On these examples, we show that the neural network potentials can be in practice decomposed in relevance contributions to different orders, that can be directly interpreted and provide physical insights on the systems of interest.

The rapid advancement of Generative AI (GenAI) technologies offers transformative opportunities within Australia's critical technologies of national interest while introducing unique security challenges. This paper presents SecGenAI, a comprehensive security framework for cloud-based GenAI applications, with a focus on Retrieval-Augmented Generation (RAG) systems. SecGenAI addresses functional, infrastructure, and governance requirements, integrating end-to-end security analysis to generate specifications emphasizing data privacy, secure deployment, and shared responsibility models. Aligned with Australian Privacy Principles, AI Ethics Principles, and guidelines from the Australian Cyber Security Centre and Digital Transformation Agency, SecGenAI mitigates threats such as data leakage, adversarial attacks, and model inversion. The framework's novel approach combines advanced machine learning techniques with robust security measures, ensuring compliance with Australian regulations while enhancing the reliability and trustworthiness of GenAI systems. This research contributes to the field of intelligent systems by providing actionable strategies for secure GenAI implementation in industry, fostering innovation in AI applications, and safeguarding national interests.

Recent legislative regulations have underlined the need for accountable and transparent artificial intelligence systems and have contributed to a growing interest in the Explainable Artificial Intelligence (XAI) field. Nonetheless, the lack of standardized criteria to validate explanation methodologies remains a major obstacle to developing trustworthy systems. We address a crucial yet often overlooked aspect of XAI, the robustness of explanations, which plays a central role in ensuring trust in both the system and the provided explanation. To this end, we propose a novel approach to analyse the robustness of neural network explanations to non-adversarial perturbations, leveraging the manifold hypothesis to produce new perturbed datapoints that resemble the observed data distribution. We additionally present an ensemble method to aggregate various explanations, showing how merging explanations can be beneficial for both understanding the model's decision and evaluating the robustness. The aim of our work is to provide practitioners with a framework for evaluating the trustworthiness of model explanations. Experimental results on feature importances derived from neural networks applied to tabular datasets highlight the importance of robust explanations in practical applications.

This study focuses on analyzing dermoscopy images to determine the depth of melanomas, which is a critical factor in diagnosing and treating skin cancer. The Breslow depth, measured from the top of the granular layer to the deepest point of tumor invasion, serves as a crucial parameter for staging melanoma and guiding treatment decisions. This research aims to improve the prediction of the depth of melanoma through the use of machine learning models, specifically deep learning, while also providing an analysis of the possible existance of graduation in the images characteristics which correlates with the depth of the melanomas. Various datasets, including ISIC and private collections, were used, comprising a total of 1162 images. The datasets were combined and balanced to ensure robust model training. The study utilized pre-trained Convolutional Neural Networks (CNNs). Results indicated that the models achieved significant improvements over previous methods. Additionally, the study conducted a correlation analysis between model's predictions and actual melanoma thickness, revealing a moderate correlation that improves with higher thickness values. Explainability methods such as feature visualization through Principal Component Analysis (PCA) demonstrated the capability of deep features to distinguish between different depths of melanoma, providing insight into the data distribution and model behavior. In summary, this research presents a dual contribution: enhancing the state-of-the-art classification results through advanced training techniques and offering a detailed analysis of the data and model behavior to better understand the relationship between dermoscopy images and melanoma thickness.

The last decade of computer vision has been dominated by Deep Learning architectures, thanks to their unparalleled success. Their performance, however, often comes at the cost of explainability owing to their highly non-linear nature. Consequently, a parallel field of eXplainable Artificial Intelligence (XAI) has developed with the aim of generating insights regarding the decision making process of deep learning models. An important problem in XAI is that of the generation of saliency maps. These are regions in an input image which contributed most towards the model's final decision. Most work in this regard, however, has been focused on image classification, and image segmentation - despite being a ubiquitous task - has not received the same attention. In the present work, we propose MiSuRe (Minimally Sufficient Region) as an algorithm to generate saliency maps for image segmentation. The goal of the saliency maps generated by MiSuRe is to get rid of irrelevant regions, and only highlight those regions in the input image which are crucial to the image segmentation decision. We perform our analysis on 3 datasets: Triangle (artificially constructed), COCO-2017 (natural images), and the Synapse multi-organ (medical images). Additionally, we identify a potential usecase of these post-hoc saliency maps in order to perform post-hoc reliability of the segmentation model.

Design and evaluation of AI copilots -- case studies of retail copilot templates

Michal Furmakiewicz, Chang Liu, Angus Taylor +1

Building a successful AI copilot requires a systematic approach. This paper is divided into two sections, covering the design and evaluation of a copilot respectively. A case study of developing copilot templates for the retail domain by Microsoft is used to illustrate the role and importance of each aspect. The first section explores the key technical components of a copilot's architecture, including the LLM, plugins for knowledge retrieval and actions, orchestration, system prompts, and responsible AI guardrails. The second section discusses testing and evaluation as a principled way to promote desired outcomes and manage unintended consequences when using AI in a business context. We discuss how to measure and improve its quality and safety, through the lens of an end-to-end human-AI decision loop framework. By providing insights into the anatomy of a copilot and the critical aspects of testing and evaluation, this paper provides concrete evidence of how good design and evaluation practices are essential for building effective, human-centered AI assistants.

Navigating the Shadows: Unveiling Effective Disturbances for Modern AI Content Detectors

Ying Zhou, Ben He, Le Sun

With the launch of ChatGPT, large language models (LLMs) have attracted global attention. In the realm of article writing, LLMs have witnessed extensive utilization, giving rise to concerns related to intellectual property protection, personal privacy, and academic integrity. In response, AI-text detection has emerged to distinguish between human and machine-generated content. However, recent research indicates that these detection systems often lack robustness and struggle to effectively differentiate perturbed texts. Currently, there is a lack of systematic evaluations regarding detection performance in real-world applications, and a comprehensive examination of perturbation techniques and detector robustness is also absent. To bridge this gap, our work simulates real-world scenarios in both informal and professional writing, exploring the out-of-the-box performance of current detectors. Additionally, we have constructed 12 black-box text perturbation methods to assess the robustness of current detection models across various perturbation granularities. Furthermore, through adversarial learning experiments, we investigate the impact of perturbation data augmentation on the robustness of AI-text detectors. We have released our code and data at https://github.com/zhouying20/ai-text-detector-evaluation.

Fairness-aware domain generalization (FairDG) has emerged as a critical challenge for deploying trustworthy AI systems, particularly in scenarios involving distribution shifts. Traditional methods for addressing fairness have failed in domain generalization due to their lack of consideration for distribution shifts. Although disentanglement has been used to tackle FairDG, it is limited by its strong assumptions. To overcome these limitations, we propose Fairness-aware Classifier-Guided Score-based Diffusion Models (FADE) as a novel approach to effectively address the FairDG issue. Specifically, we first pre-train a score-based diffusion model (SDM) and two classifiers to equip the model with strong generalization capabilities across different domains. Then, we guide the SDM using these pre-trained classifiers to effectively eliminate sensitive information from the generated data. Finally, the generated fair data is used to train downstream classifiers, ensuring robust performance under new data distributions. Extensive experiments on three real-world datasets demonstrate that FADE not only enhances fairness but also improves accuracy in the presence of distribution shifts. Additionally, FADE outperforms existing methods in achieving the best accuracy-fairness trade-offs.

Model Interpretation and Explainability: Towards Creating Transparency in Prediction Models

Donald Kridel, Jacob Dineen, Daniel Dolk +1

Explainable AI (XAI) has a counterpart in analytical modeling which we refer to as model explainability. We tackle the issue of model explainability in the context of prediction models. We analyze a dataset of loans from a credit card company and apply three stages: execute and compare four different prediction methods, apply the best known explainability techniques in the current literature to the model training sets to identify feature importance (FI) (static case), and finally to cross-check whether the FI set holds up under what if prediction scenarios for continuous and categorical variables (dynamic case). We found inconsistency in FI identification between the static and dynamic cases. We summarize the state of the art in model explainability and suggest further research to advance the field.