By Vincent Granville Ph.D, published in September 2022. The book is available here. For my upcoming course based on this book, see here. This book covers the foundations of machine learning, with modern approaches to solving complex problems. Emphasis is on scalability, automation, testing, optimizing, and interpretability (explainable AI).

Concept induction, which is based on formal logical reasoning over description logics, has been used in ontology engineering in order to create ontology (TBox) axioms from the base data (ABox) graph. In this paper, we show that it can also be used to explain data differentials, for example in the context of Explainable AI (XAI), and we show that it can in fact be done in a way that is meaningful to a human observer.

This paper establishes a dual theory about knowledge and argumentation. Our idea is rooted at both epistemic logic and argumentation theory, and we aim to merge these two fields, not just in a superficial way but to thoroughly disclose the intrinsic relevance between knowledge and argumentation. Specifically, we define epistemic Kripke models and argument Kripke models as a dual pair, and then work out a two-way generation method between these two types of Kripke models. Such generation is rigorously justified by a duality theorem on modal formulae's invariance. We also provide realistic examples to demonstrate our generation, through which our framework's practical utility gets strongly advocated. We finally propose a philosophical thesis that knowledge is essentially dynamic, and we draw certain connection to Maxwell's demon as well as the well-known proverb "knowledge is power".

Counterfactual explanations (CEs) are a powerful means for understanding how decisions made by algorithms can be changed. Researchers have proposed a number of desiderata that CEs should meet to be practically useful, such as requiring minimal effort to enact, or complying with causal models. We consider a further aspect to improve the usability of CEs: robustness to adverse perturbations, which may naturally happen due to unfortunate circumstances. Since CEs typically prescribe a sparse form of intervention (i.e., only a subset of the features should be changed), we study the effect of addressing robustness separately for the features that are recommended to be changed and those that are not. Our definitions are workable in that they can be incorporated as penalty terms in the loss functions that are used for discovering CEs. To experiment with robustness, we create and release code where five data sets (commonly used in the field of fair and explainable machine learning) have been enriched with feature-specific annotations that can be used to sample meaningful perturbations. Our experiments show that CEs are often not robust and, if adverse perturbations take place (even if not worst-case), the intervention they prescribe may require a much larger cost than anticipated, or even become impossible. However, accounting for robustness in the search process, which can be done rather easily, allows discovering robust CEs systematically. Robust CEs make additional intervention to contrast perturbations much less costly than non-robust CEs. We also find that robustness is easier to achieve for the features to change, posing an important point of consideration for the choice of what counterfactual explanation is best for the user. Our code is available at: https://github.com/marcovirgolin/robust-counterfactuals.

In the wake of the "Great Reshuffle," companies continue to reevaluate their approach to recruitment and retention. In order to drive efficiency and remain effective at scale, business leaders are increasingly turning to new technologies for support. One of the most valuable technologies supporting talent management strategies today is artificial intelligence (AI). It has the potential to revolutionize the way in which businesses interact with the wider talent landscape, helping HR teams and recruiters fill much-needed positions and identify the skill sets in most demand. A critical, but often overlooked, element for talent stakeholders to take into account is ensuring all candidates--internal or external--are given fair and equal consideration to the open opportunities.

Abakar Saidov is co-founder and CEO of Beamery, a leader in talent lifecycle management. In the wake of the "Great Reshuffle," companies continue to reevaluate their approach to recruitment and retention. In order to drive efficiency and remain effective at scale, business leaders are increasingly turning to new technologies for support. One of the most valuable technologies supporting talent management strategies today is artificial intelligence (AI). It has the potential to revolutionize the way in which businesses interact with the wider talent landscape, helping HR teams and recruiters fill much-needed positions and identify the skill sets in most demand.

The goal of this work is to develop a robot equipped with goal-driven explainability, i.e. a robot will explain its behavior to achieve a particular goal in a collaborative setting. The major factor in goal-driven explainability is the human'theory of mind'. In this work, we will employ Leslies' theory of mind model which includes a mechanical agency, an actionable agency and a belief agency. This thesis will focus on explaining the desire of the robot and the belief of the human if its different to the robot's intention or desire. We aim to develop a common theoretical framework for the development of understandable robots which will include learning to generate explanations, non-verbal and verbal ways of communication and explanations in context.

Recent years have seen a surge of interest in the field of explainable AI (XAI), with a plethora of algorithms proposed in the literature. However, a lack of consensus on how to evaluate XAI hinders the advancement of the field. We highlight that XAI is not a monolithic set of technologies -- researchers and practitioners have begun to leverage XAI algorithms to build XAI systems that serve different usage contexts, such as model debugging and decision-support. Algorithmic research of XAI, however, often does not account for these diverse downstream usage contexts, resulting in limited effectiveness or even unintended consequences for actual users, as well as difficulties for practitioners to make technical choices. We argue that one way to close the gap is to develop evaluation methods that account for different user requirements in these usage contexts. Towards this goal, we introduce a perspective of contextualized XAI evaluation by considering the relative importance of XAI evaluation criteria for prototypical usage contexts of XAI. To explore the context dependency of XAI evaluation criteria, we conduct two survey studies, one with XAI topical experts and another with crowd workers. Our results urge for responsible AI research with usage-informed evaluation practices, and provide a nuanced understanding of user requirements for XAI in different usage contexts.

Ontology is a popular method for knowledge representation in different domains, including the legal domain, and description logics (DL) is commonly used as its description language. To handle reasoning based on inconsistent DL-based legal ontologies, the current paper presents a structured argumentation framework particularly for reasoning in legal contexts on the basis of ASPIC+, and translates the legal ontology into formulas and rules of an argumentation theory. With a particular focus on the design of autonomous vehicles from the perspective of legal AI, we show that using this combined theory of formal argumentation and DL-based legal ontology, acceptable assertions can be obtained based on inconsistent ontologies, and the traditional reasoning tasks of DL ontologies can also be accomplished. In addition, a formal definition of explanations for the result of reasoning is presented.

In the last few years, many explanation methods based on the perturbations of input data have been introduced to improve our understanding of decisions made by black-box models. The goal of this work is to introduce a novel perturbation scheme so that more faithful and robust explanations can be obtained. Our study focuses on the impact of perturbing directions on the data topology. We show that perturbing along the orthogonal directions of the input manifold better preserves the data topology, both in the worst-case analysis of the discrete Gromov-Hausdorff distance and in the average-case analysis via persistent homology. From those results, we introduce EMaP algorithm, realizing the orthogonal perturbation scheme. Our experiments show that EMaP not only improves the explainers' performance but also helps them overcome a recently-developed attack against perturbation-based methods.