Generalizing the attack structure in argumentation frameworks (AFs) has been studied in different ways. Most prominently, the binary attack relation of Dung frameworks has been extended to the notion of collective attacks. The resulting formalism is often termed SETAFs. Another approach is provided via abstract dialectical frameworks (ADFs), where acceptance conditions specify the relation between arguments; restricting these conditions naturally allows for so-called support-free ADFs. The aim of the paper is to shed light on the relation between these two different approaches. To this end, we investigate and compare the expressiveness of SETAFs and support-free ADFs under the lens of 3-valued semantics. Our results show that it is only the presence of unsatisfiable acceptance conditions in support-free ADFs that discriminate the two approaches.

THuMP is a multi-disciplinary project, with the ambitious goal of advancing the state-of-the-art in trustworthy human-AI decision-support systems. ThUMP will address the technical challenges involved in creating explainable AI (XAI) systems, with a focus on Visualization for Explainable Planning and Argumentation, so that people using the system can better understand the rationale behind and trust suggestions made by an AI system. This project is conducted in collaboration with three project partners: Schlumberger and Save the Children, which provide use cases for the project, and a law firm whowill cooperate in considering legal implications of enhancing machines with transparency and the ability to explain. The candidate will be responsible for conducting research around the interfaces required to support explainability in the context of decision making in human-machine partnerships. Tasks will involve designing new visual layouts, building the interaction infrastructure for the project, developing a prototype interface for communicating with users, designing and conducting experiments with human subjects based on the use cases that will be co-created with the project partners.

Explainability is one of the key elements for building trust in AI systems. Among numerous attempts to make AI explainable, quantifying the effect of explanations remains a challenge in conducting human-AI collaborative tasks. Aside from the ability to predict the overall behavior of AI, in many applications, users need to understand an AI agent's competency in different aspects of the task domain. In this paper, we evaluate the impact of explanations on the user's mental model of AI agent competency within the task of visual question answering (VQA). We quantify users' understanding of competency, based on the correlation between the actual system performance and user rankings. We introduce an explainable VQA system that uses spatial and object features and is powered by the BERT language model. Each group of users sees only one kind of explanation to rank the competencies of the VQA model. The proposed model is evaluated through between-subject experiments to probe explanations' impact on the user's perception of competency. The comparison between two VQA models shows BERT based explanations and the use of object features improve the user's prediction of the model's competencies.

Deep learning bears promise for drug discovery, including advanced image analysis, prediction of molecular structure and function, and automated generation of innovative chemical entities with bespoke properties. Despite the growing number of successful prospective applications, the underlying mathematical models often remain elusive to interpretation by the human mind. There is a demand for 'explainable' deep learning methods to address the need for a new narrative of the machine language of the molecular sciences. This review summarizes the most prominent algorithmic concepts of explainable artificial intelligence, and dares a forecast of the future opportunities, potential applications, and remaining challenges.

Since AI involves interactions between machines and humans--rather than just the former replacing the latter--'explainable AI' is a new challenge. Intelligent systems, based on machine learning, are penetrating many aspects of our society. They span a large variety of applications--from the seemingly harmless automation of micro-tasks, such as the suggestion of synonymous phrases in text editors, to more contestable uses, such as in jail-or-release decisions, anticipating child-services interventions, predictive policing and many others. Researchers have shown that for some tasks, such as lung-cancer screening, intelligent systems are capable of outperforming humans. In many other cases, however, they have not lived up to exaggerated expectations.

A method for counterfactual explanation of machine learning survival models is proposed. One of the difficulties of solving the counterfactual explanation problem is that the classes of examples are implicitly defined through outcomes of a machine learning survival model in the form of survival functions. A condition that establishes the difference between survival functions of the original example and the counterfactual is introduced. This condition is based on using a distance between mean times to event. It is shown that the counterfactual explanation problem can be reduced to a standard convex optimization problem with linear constraints when the explained black-box model is the Cox model. For other black-box models, it is proposed to apply the well-known Particle Swarm Optimization algorithm. A lot of numerical experiments with real and synthetic data demonstrate the proposed method.

Counterfactual explanations are one of the most popular methods to make predictions of black box machine learning models interpretable by providing explanations in the form of `what-if scenarios'. Most current approaches optimize a collapsed, weighted sum of multiple objectives, which are naturally difficult to balance a-priori. We propose the Multi-Objective Counterfactuals (MOC) method, which translates the counterfactual search into a multi-objective optimization problem. Our approach not only returns a diverse set of counterfactuals with different trade-offs between the proposed objectives, but also maintains diversity in feature space. This enables a more detailed post-hoc analysis to facilitate better understanding and also more options for actionable user responses to change the predicted outcome. Our approach is also model-agnostic and works for numerical and categorical input features. We show the usefulness of MOC in concrete cases and compare our approach with state-of-the-art methods for counterfactual explanations.

Counterfactual explanations are usually obtained by identifying the smallest change made to an input to change a prediction made by a fixed model (hereafter called sparse methods). Recent work, however, has revitalized an old insight: there often does not exist one superior solution to a prediction problem with respect to commonly used measures of interest (e.g. error rate). In fact, often multiple different classifiers give almost equal solutions. This phenomenon is known as predictive multiplicity (Breiman, 2001; Marx et al., 2019). In this work, we derive a general upper bound for the costs of counterfactual explanations under predictive multiplicity. Most notably, it depends on a discrepancy notion between two classifiers, which describes how differently they treat negatively predicted individuals. We then compare sparse and data support approaches empirically on real-world data. The results show that data support methods are more robust to multiplicity of different models. At the same time, we show that those methods have provably higher cost of generating counterfactual explanations under one fixed model. In summary, our theoretical and empiricaln results challenge the commonly held view that counterfactual recommendations should be sparse in general.

The notion of concept drift refers to the phenomenon that the distribution, which is underlying the observed data, changes over time; as a consequence machine learning models may become inaccurate and need adjustment. While there do exist methods to detect concept drift or to adjust models in the presence of observed drift, the question of explaining drift has hardly been considered so far. This problem is of importance, since it enables an inspection of the most prominent features where drift manifests itself; hence it enables human understanding of the necessity of change and it increases acceptance of life-long learning models. In this paper we present a novel technology, which characterizes concept drift in terms of the characteristic change of spatial features represented by typical examples based on counterfactual explanations. We establish a formal definition of this problem, derive an efficient algorithmic solution based on counterfactual explanations, and demonstrate its usefulness in several examples.

Machine learning becomes increasingly important to tune or even synthesize the behavior of safety-critical components in highly non-trivial environments, where the inability to understand learned components in general, and neural nets in particular, poses serious obstacles to their adoption. Explainability and interpretability methods for learned systems have gained considerable academic attention, but the focus of current approaches on only one aspect of explanation, at a fixed level of abstraction, and limited if any formal guarantees, prevents those explanations from being digestible by the relevant stakeholders (e.g., end users, certification authorities, engineers) with their diverse backgrounds and situation-specific needs. We introduce Fanoos, a flexible framework for combining formal verification techniques, heuristic search, and user interaction to explore explanations at the desired level of granularity and fidelity. We demonstrate the ability of Fanoos to produce and adjust the abstractness of explanations in response to user requests on a learned controller for an inverted double pendulum and on a learned CPU usage model.