The field of eXplainable Artificial Intelligence (XAI) aims to bring transparency to today's powerful but opaque deep learning models. While local XAI methods explain individual predictions in form of attribution maps, thereby identifying where important features occur (but not providing information about what they represent), global explanation techniques visualize what concepts a model has generally learned to encode. Both types of methods thus only provide partial insights and leave the burden of interpreting the model's reasoning to the user. In this work we introduce the Concept Relevance Propagation (CRP) approach, which combines the local and global perspectives and thus allows answering both the "where" and "what" questions for individual predictions. We demonstrate the capability of our method in various settings, showcasing that CRP leads to more human interpretable explanations and provides deep insights into the model's representation and reasoning through concept atlases, concept composition analyses, and quantitative investigations of concept subspaces and their role in fine-grained decision making.

Explainable recommender systems can explain their recommendation decisions, enhancing user trust in the systems. Most explainable recommender systems either rely on human-annotated rationales to train models for explanation generation or leverage the attention mechanism to extract important text spans from reviews as explanations. The extracted rationales are often confined to an individual review and may fail to identify the implicit features beyond the review text. To avoid the expensive human annotation process and to generate explanations beyond individual reviews, we propose to incorporate a geometric prior learnt from user-item interactions into a variational network which infers latent factors from user-item reviews. The latent factors from an individual user-item pair can be used for both recommendation and explanation generation, which naturally inherit the global characteristics encoded in the prior knowledge. Experimental results on three e-commerce datasets show that our model significantly improves the interpretability of a variational recommender using the Wasserstein distance while achieving performance comparable to existing content-based recommender systems in terms of recommendation behaviours.

Model Multiplicity (MM) arises when multiple, equally performing machine learning models can be trained to solve the same prediction task. Recent studies show that models obtained under MM may produce inconsistent predictions for the same input. When this occurs, it becomes challenging to provide counterfactual explanations (CEs), a common means for offering recourse recommendations to individuals negatively affected by models' predictions. In this paper, we formalise this problem, which we name recourse-aware ensembling, and identify several desirable properties which methods for solving it should satisfy. We show that existing ensembling methods, naturally extended in different ways to provide CEs, fail to satisfy these properties. We then introduce argumentative ensembling, deploying computational argumentation to guarantee robustness of CEs to MM, while also accommodating customisable user preferences. We show theoretically and experimentally that argumentative ensembling satisfies properties which the existing methods lack, and that the trade-offs are minimal wrt accuracy.

While most prior research into the universe of political discourses is based in the genres of debate and speeches, studies of academic political discourse have been sparse. One of the goals of the project SKILL, from which this paper stems, is to fill this gap. SKILL - A social science lab for research-based learning - is dedicated to building and applying AI technologies to facilitate analysis of argumentation in scholarly articles in political science, especially in the context of teaching International Relations (IR). The ultimate goal of SKILL is to provide students with AI tools which would facilitate comprehension of original articles used as part of teaching syllabi and which would coach them in producing expert argumentation in the field. In order to gain insight into the structure and properties of arguments in the domain of political science theory, we developed an annotation scheme which enables analysis of scholarly IR discourse in terms of interaction between argumentation and types of domain content contributing to arguments. The scheme comprises two orthogonal dimensions: discourse and content domain.

In order to fully harness the potential of machine learning, it is crucial to establish a system that renders the field more accessible and less daunting for individuals who may not possess a comprehensive understanding of its intricacies. The paper describes the design of a system that integrates AutoML, XAI, and synthetic data generation to provide a great UX design for users. The system allows users to navigate and harness the power of machine learning while abstracting its complexities and providing high usability. The paper proposes two novel classifiers, Logistic Regression Forest and Support Vector Tree, for enhanced model performance, achieving 96\% accuracy on a diabetes dataset and 93\% on a survey dataset. The paper also introduces a model-dependent local interpreter called MEDLEY and evaluates its interpretation against LIME, Greedy, and Parzen. Additionally, the paper introduces LLM-based synthetic data generation, library-based data generation, and enhancing the original dataset with GAN. The findings on synthetic data suggest that enhancing the original dataset with GAN is the most reliable way to generate synthetic data, as evidenced by KS tests, standard deviation, and feature importance. The authors also found that GAN works best for quantitative datasets.

This paper presents a systematic literature review (SLR) on the explainability and interpretability of machine learning (ML) models within the context of predictive process mining, using the PRISMA framework. Given the rapid advancement of artificial intelligence (AI) and ML systems, understanding the "black-box" nature of these technologies has become increasingly critical. Focusing specifically on the domain of process mining, this paper delves into the challenges of interpreting ML models trained with complex business process data. We differentiate between intrinsically interpretable models and those that require post-hoc explanation techniques, providing a comprehensive overview of the current methodologies and their applications across various application domains. Through a rigorous bibliographic analysis, this research offers a detailed synthesis of the state of explainability and interpretability in predictive process mining, identifying key trends, challenges, and future directions. Our findings aim to equip researchers and practitioners with a deeper understanding of how to develop and implement more trustworthy, transparent, and effective intelligent systems for predictive process analytics.

Machine learning techniques are increasingly used for high-stakes decision-making, such as college admissions, loan attribution or recidivism prediction. Thus, it is crucial to ensure that the models learnt can be audited or understood by human users, do not create or reproduce discrimination or bias, and do not leak sensitive information regarding their training data. Indeed, interpretability, fairness and privacy are key requirements for the development of responsible machine learning, and all three have been studied extensively during the last decade. However, they were mainly considered in isolation, while in practice they interplay with each other, either positively or negatively. In this Systematization of Knowledge (SoK) paper, we survey the literature on the interactions between these three desiderata. More precisely, for each pairwise interaction, we summarize the identified synergies and tensions. These findings highlight several fundamental theoretical and empirical conflicts, while also demonstrating that jointly considering these different requirements is challenging when one aims at preserving a high level of utility. To solve this issue, we also discuss possible conciliation mechanisms, showing that a careful design can enable to successfully handle these different concerns in practice.

We investigate interaction patterns for humans interacting with explainable and non-explainable robots. Non-explainable robots are here robots that do not explain their actions or non-actions, neither do they give any other feedback during interaction, in contrast to explainable robots. We video recorded and analyzed human behavior during a board game, where 20 humans verbally instructed either an explainable or non-explainable Pepper robot to move objects on the board. The transcriptions and annotations of the videos were transformed into transactions for association rule mining. Association rules discovered communication patterns in the interaction between the robots and the humans, and the most interesting rules were also tested with regular chi-square tests. Some statistically significant results are that there is a strong correlation between men and non-explainable robots and women and explainable robots, and that humans mirror some of the robot's modality. Our results also show that it is important to contextualize human interaction patterns, and that this can be easily done using association rules as an investigative tool. The presented results are important when designing robots that should adapt their behavior to become understandable for the interacting humans.

Generating counterfactual explanations is one of the most effective approaches for uncovering the inner workings of black-box neural network models and building user trust. While remarkable strides have been made in generative modeling using diffusion models in domains like vision, their utility in generating counterfactual explanations in structured modalities remains unexplored. In this paper, we introduce Structured Counterfactual Diffuser or SCD, the first plug-and-play framework leveraging diffusion for generating counterfactual explanations in structured data. SCD learns the underlying data distribution via a diffusion model which is then guided at test time to generate counterfactuals for any arbitrary black-box model, input, and desired prediction. Our experiments show that our counterfactuals not only exhibit high plausibility compared to the existing state-of-the-art but also show significantly better proximity and diversity.

We argue that interpretations of machine learning (ML) models or the model-building process can bee seen as a form of sensitivity analysis (SA), a general methodology used to explain complex systems in many fields such as environmental modeling, engineering, or economics. We address both researchers and practitioners, calling attention to the benefits of a unified SA-based view of explanations in ML and the necessity to fully credit related work. We bridge the gap between both fields by formally describing how (a) the ML process is a system suitable for SA, (b) how existing ML interpretation methods relate to this perspective, and (c) how other SA techniques could be applied to ML.