Explainable artificial intelligence (XAI) is a set of processes and methods that allows human users to comprehend and trust the results and output created by machine learning algorithms. Explainable AI is used to describe an AI model, its expected impact and potential biases. It helps characterize model accuracy, fairness, transparency and outcomes in AI-powered decision making. SHAP (SHapley Additive exPlanations) is a framework that explains the output of any model using Shapley values, a game-theoretic approach often used for optimal credit allocation. While this can be used on any black-box model, SHAP can compute more efficiently on specific model classes (like tree ensembles).

Can your brand explain how its artificial intelligence (AI) applications work, and why they make the decisions they do? Brand trust is hard to win and easy to lose, and transparent and easily explainable AI applications are a great start towards building customers' trust and enhancing the efficiency and effectiveness of AI apps. This article looks at Explainable AI (XAI), and why it should be a part of your brand's AI strategy. Typical AI apps are often referred to as "black box" AI because whatever occurs within the application is relatively unknown to all but those data scientists, programmers and designers who created it. Individually, even those people may not be able to explain anything outside of their primary domain.

More than two dozen artificial intelligence experts from business and academia, including Texas McCombs, explored the importance of understanding how machine learning systems arrive at their conclusions so humans can trust those results. Although AI is more than 50 years old, "deep learning has been a mini-scientific revolution" since the 2010s, said one keynote speaker, Charles Elkan, a professor of computer science at the University of California, San Diego. Alice Xiang, a lawyer and a senior research scientist for Sony Group, said, "I see explainability as an important part of providing transparency and, in turn, enabling accountability." She noted the challenge of black boxes, citing as examples drug-sniffing dogs, whose abilities are mysterious but highly accurate, and the horse Clever Hans, who appeared to understand math but was really following cues from its owner. In a panel discussion called "Adopting AI," James Guszcza, a behavioral research affiliate at Stanford University and chief data scientist on leave from Deloitte LLP, said: "I think one of the previous speakers said we need to be interdisciplinary; I take it a little bit further and say we need to be transdisciplinary."

While machine learning models are usually assumed to always output a prediction, there also exist extensions in the form of reject options which allow the model to reject inputs where only a prediction with an unacceptably low certainty would be possible. With the ongoing rise of eXplainable AI, a lot of methods for explaining model predictions have been developed. However, understanding why a given input was rejected, instead of being classified by the model, is also of interest. Surprisingly, explanations of rejects have not been considered so far. We propose to use counterfactual explanations for explaining rejects and investigate how to efficiently compute counterfactual explanations of different reject options for an important class of models, namely prototype-based classifiers such as learning vector quantization models.

Accurate electricity demand forecasts play a crucial role in sustainable power systems. To enable better decision-making especially for demand flexibility of the end-user, it is necessary to provide not only accurate but also understandable and actionable forecasts. To provide accurate forecasts Global Forecasting Models (GFM) trained across time series have shown superior results in many demand forecasting competitions and real-world applications recently, compared with univariate forecasting approaches. We aim to fill the gap between the accuracy and the interpretability in global forecasting approaches. In order to explain the global model forecasts, we propose Local Interpretable Model-agnostic Rule-based Explanations for Forecasting (LIMREF), a local explainer framework that produces k-optimal impact rules for a particular forecast, considering the global forecasting model as a black-box model, in a model-agnostic way. It provides different types of rules that explain the forecast of the global model and the counterfactual rules, which provide actionable insights for potential changes to obtain different outputs for given instances. We conduct experiments using a large-scale electricity demand dataset with exogenous features such as temperature and calendar effects. Here, we evaluate the quality of the explanations produced by the LIMREF framework in terms of both qualitative and quantitative aspects such as accuracy, fidelity, and comprehensibility and benchmark those against other local explainers.

This paper provides new theory to support to the eXplainable AI (XAI) method Contextual Importance and Utility (CIU). CIU arithmetic is based on the concepts of Multi-Attribute Utility Theory, which gives CIU a solid theoretical foundation. The novel concept of contextual influence is also defined, which makes it possible to compare CIU directly with so-called additive feature attribution (AFA) methods for model-agnostic outcome explanation. One key takeaway is that the "influence" concept used by AFA methods is inadequate for outcome explanation purposes even for simple models to explain. Experiments with simple models show that explanations using contextual importance (CI) and contextual utility (CU) produce explanations where influence-based methods fail. It is also shown that CI and CU guarantees explanation faithfulness towards the explained model.

Many existing methods of counterfactual explanations ignore the intrinsic relationships between data attributes and thus fail to generate realistic counterfactuals. Moreover, the existing methods that account for relationships between data attributes require domain knowledge, which limits their applicability in complex real-world applications. In this paper, we propose a novel approach to realistic counterfactual explanations that preserve relationships between data attributes. The model directly learns the relationships by a variational auto-encoder without domain knowledge and then learns to disturb the latent space accordingly. We conduct extensive experiments on both synthetic and real-world datasets. The results demonstrate that the proposed method learns relationships from the data and preserves these relationships in generated counterfactuals.

Although standard Machine Learning models are optimized for making predictions about observations, more and more they are used for making predictions about the results of actions. An important goal of Explainable Artificial Intelligence (XAI) is to compensate for this mismatch by offering explanations about the predictions of an ML-model which ensure that they are reliably action-guiding. As action-guiding explanations are causal explanations, the literature on this topic is starting to embrace insights from the literature on causal models. Here I take a step further down this path by formally defining the causal notions of sufficient explanations and counterfactual explanations. I show how these notions relate to (and improve upon) existing work, and motivate their adequacy by illustrating how different explanations are action-guiding under different circumstances. Moreover, this work is the first to offer a formal definition of actual causation that is founded entirely in action-guiding explanations. Although the definitions are motivated by a focus on XAI, the analysis of causal explanation and actual causation applies in general. I also touch upon the significance of this work for fairness in AI by showing how actual causation can be used to improve the idea of path-specific counterfactual fairness.

This paper deals with the problem of finding the preferred extensions of an argumentation framework by means of a bijection with the naive extensions of another framework. First we consider the case where an argumentation framework is naive-realizable: its naive and preferred extensions are equal. Recognizing naive-realizable argumentation frameworks is hard, but we show that it is tractable for frameworks with bounded in-degree. Next, we give a bijection between the preferred extensions of an argumentation framework being admissible-closed (the intersection of two admissible sets is admissible) and the naive extensions of another framework on the same set of arguments. On the other hand, we prove that identifying admissible-closed argumentation frameworks is coNP-complete. At last, we introduce the notion of irreducible self-defending sets as those that are not the union of others. It turns out there exists a bijection between the preferred extensions of an argumentation framework and the naive extensions of a framework on its irreducible self-defending sets. Consequently, the preferred extensions of argumentation frameworks with some lattice properties can be listed with polynomial delay and polynomial space.

In the future, AI will increasingly find its way into systems that can potentially cause physical harm to humans. For such safety-critical systems, it must be demonstrated that their residual risk does not exceed what is acceptable. This includes, in particular, the AI components that are part of such systems' safety-related functions. Assurance cases are an intensively discussed option today for specifying a sound and comprehensive safety argument to demonstrate a system's safety. In previous work, it has been suggested to argue safety for AI components by structuring assurance cases based on two complementary risk acceptance criteria. One of these criteria is used to derive quantitative targets regarding the AI. The argumentation structures commonly proposed to show the achievement of such quantitative targets, however, focus on failure rates from statistical testing. Further important aspects are only considered in a qualitative manner -- if at all. In contrast, this paper proposes a more holistic argumentation structure for having achieved the target, namely a structure that integrates test results with runtime aspects and the impact of scope compliance and test data quality in a quantitative manner. We elaborate different argumentation options, present the underlying mathematical considerations, and discuss resulting implications for their practical application. Using the proposed argumentation structure might not only increase the integrity of assurance cases but may also allow claims on quantitative targets that would not be justifiable otherwise.