There has been a recent resurgence of interest in explainable artificial intelligence (XAI) that aims to reduce the opaqueness of AI-based decision-making systems, allowing humans to scrutinize and trust them. Prior work in this context has focused on the attribution of responsibility for an algorithm's decisions to its inputs wherein responsibility is typically approached as a purely associational concept. In this paper, we propose a principled causality-based approach for explaining black-box decision-making systems that addresses limitations of existing methods in XAI. At the core of our framework lies probabilistic contrastive counterfactuals, a concept that can be traced back to philosophical, cognitive, and social foundations of theories on how humans generate and select explanations. We show how such counterfactuals can quantify the direct and indirect influences of a variable on decisions made by an algorithm, and provide actionable recourse for individuals negatively affected by the algorithm's decision.

These are heady days in data science and machine learning (DSML) according to Gartner, which identified a "glut" of innovation occurring in the market for DSML platforms. From established companies chasing AutoML or model governance to startups focusing on MLops or explainable AI, a plethora of vendors are simultaneously moving in all directions with their products as they seek to differentiate themselves amid a very diverse audience. "The DSML market is simultaneously more vibrant and messier than ever," a gaggle of Gartner analysts led by Peter Krensky wrote in the Magic Quadrant for DSML Platforms, which was published earlier this month. "The definitions and parameters of data science and data scientists continue to evolve, and the market is dramatically different from how it was in 2014, when we published the first Magic Quadrant on it." The 2021 Magic Quadrant for DSML is heavily represented by companies to the right of the axis, which anybody who's familiar with Gartner's quadrant-based assessment method knows represents the "completeness of vision."

Explainable AI (XAI) principles are a set of guidelines for the fundamental properties that explainable AI systems should adopt. Explainable AI seeks to explain the way that AI systems work. These four principles capture a variety of disciplines that contribute to explainable AI, including computer science, engineering and psychology. The four explainable AI principles apply individually, so the presence of one does not imply that others will be present. The NIST suggests that each principle should be evaluated in its own right.

There has been a recent resurgence of interest in explainable artificial intelligence (XAI) that aims to reduce the opaqueness of AI-based decision-making systems, allowing humans to scrutinize and trust them. Prior work in this context has focused on the attribution of responsibility for an algorithm's decisions to its inputs wherein responsibility is typically approached as a purely associational concept. In this paper, we propose a principled causality-based approach for explaining black-box decision-making systems that addresses limitations of existing methods in XAI. At the core of our framework lies probabilistic contrastive counterfactuals, a concept that can be traced back to philosophical, cognitive, and social foundations of theories on how humans generate and select explanations. We show how such counterfactuals can quantify the direct and indirect influences of a variable on decisions made by an algorithm, and provide actionable recourse for individuals negatively affected by the algorithm's decision. Unlike prior work, our system, LEWIS: (1)can compute provably effective explanations and recourse at local, global and contextual levels (2)is designed to work with users with varying levels of background knowledge of the underlying causal model and (3)makes no assumptions about the internals of an algorithmic system except for the availability of its input-output data. We empirically evaluate LEWIS on three real-world datasets and show that it generates human-understandable explanations that improve upon state-of-the-art approaches in XAI, including the popular LIME and SHAP. Experiments on synthetic data further demonstrate the correctness of LEWIS's explanations and the scalability of its recourse algorithm.

Machine Learning (ML) seems to be one of the most promising solution to automate partially or completely some of the complex tasks currently realized by humans, such as driving vehicles, recognizing voice, etc. It is also an opportunity to implement and embed new capabilities out of the reach of classical implementation techniques. However, ML techniques introduce new potential risks. Therefore, they have only been applied in systems where their benefits are considered worth the increase of risk. In practice, ML techniques raise multiple challenges that could prevent their use in systems submitted to certification constraints. But what are the actual challenges? Can they be overcome by selecting appropriate ML techniques, or by adopting new engineering or certification practices? These are some of the questions addressed by the ML Certification 3 Workgroup (WG) set-up by the Institut de Recherche Technologique Saint Exup\'ery de Toulouse (IRT), as part of the DEEL Project.

As transparency becomes key for robotics and AI, it will be necessary to evaluate the methods through which transparency is provided, including automatically generated natural language (NL) explanations. Here, we explore parallels between the generation of such explanations and the much-studied field of evaluation of Natural Language Generation (NLG). Specifically, we investigate which of the NLG evaluation measures map well to explanations. We present the ExBAN corpus: a crowd-sourced corpus of NL explanations for Bayesian Networks. We run correlations comparing human subjective ratings with NLG automatic measures. We find that embedding-based automatic NLG evaluation methods, such as BERTScore and BLEURT, have a higher correlation with human ratings, compared to word-overlap metrics, such as BLEU and ROUGE. This work has implications for Explainable AI and transparent robotic and autonomous systems.

The automotive industry is seen to have witnessed an increasing level of development in the past decades; from manufacturing manually operated vehicles to manufacturing vehicles with high level of automation. With the recent developments in Artificial Intelligence (AI), automotive companies now employ high performance AI models to enable vehicles to perceive their environment and make driving decisions with little or no influence from a human. With the hope to deploy autonomous vehicles (AV) on a commercial scale, the acceptance of AV by society becomes paramount and may largely depend on their degree of transparency, trustworthiness, and compliance to regulations. The assessment of these acceptance requirements can be facilitated through the provision of explanations for AVs' behaviour. Explainability is therefore seen as an important requirement for AVs. AVs should be able to explain what they have 'seen', done and might do in environments where they operate. In this paper, we provide a comprehensive survey of the existing work in explainable autonomous driving. First, we open by providing a motivation for explanations and examining existing standards related to AVs. Second, we identify and categorise the different stakeholders involved in the development, use, and regulation of AVs and show their perceived need for explanation. Third, we provide a taxonomy of explanations and reviewed previous work on explanation in the different AV operations. Finally, we draw a close by pointing out pertinent challenges and future research directions. This survey serves to provide fundamental knowledge required of researchers who are interested in explanation in autonomous driving.

There has been a growing interest in model-agnostic methods that can make deep learning models more transparent and explainable to a user. Some researchers recently argued that for a machine to achieve a certain degree of human-level explainability, this machine needs to provide human causally understandable explanations, also known as causability. A specific class of algorithms that have the potential to provide causability are counterfactuals. This paper presents an in-depth systematic review of the diverse existing body of literature on counterfactuals and causability for explainable artificial intelligence. We performed an LDA topic modelling analysis under a PRISMA framework to find the most relevant literature articles. This analysis resulted in a novel taxonomy that considers the grounding theories of the surveyed algorithms, together with their underlying properties and applications in real-world data. This research suggests that current model-agnostic counterfactual algorithms for explainable AI are not grounded on a causal theoretical formalism and, consequently, cannot promote causability to a human decision-maker. Our findings suggest that the explanations derived from major algorithms in the literature provide spurious correlations rather than cause/effects relationships, leading to sub-optimal, erroneous or even biased explanations. This paper also advances the literature with new directions and challenges on promoting causability in model-agnostic approaches for explainable artificial intelligence.

The banking software company is teaming up with Canadian Western Bank (CWB) to provide its new Temenos Virtual COO solution to small and medium-sized businesses (SMBs). The product is built on top of Temenos' omnichannel digital banking platform and utilizes explainable AI (XAI) and analytics to support financial decision-making at SMBs. By aggregating banking and business data, SMBs are able to assess their current and projected financial health through the use of XAI-powered models that simulate different business scenarios. Banks could utilize XAI technology to rectify the black box problem associated with traditional AI models used in banking. While a powerful tool in terms of generating financial insights, banks should use XAI to complement their existing interactions with customers--not replace them.

Fairness and explainability are two important and closely related requirements of decision making systems. While fairness and explainability of decision making systems have been extensively studied independently, only little effort has been put into studying fairness of explanations on their own. Current explanations can be unfair to an individual: an example is given by counterfactual explanations which propose different actions to change the output class to two similar individuals. In this work we formally and empirically study individual fairness and its mathematical formalization as robustness for counterfactual explanations as a prominent instance of contrasting explanations. In addition, we propose to use plausible counterfactuals instead of closest counterfactuals for improving the individual fairness of counterfactual explanations.