Search

Explainable AI (XAI) aims to explain the behavior of opaque AI systems, and in this way, increase their trustworthiness. However, current XAI methods are explanatorily deficient. On the one hand, "top-down" XAI methods allow for global and local prediction, but rarely support targeted internal interventions. On the other hand, "bottom-up" XAI methods may support such interventions, but rarely provide global behavioral predictions. To overcome this limitation, we argue that XAI should follow the lead of cognitive science in developing cognitive models that simultaneously reproduce behavior and support interventions. Indeed, novel methods such as mechanistic interpretability and causal abstraction analysis already reflect cognitive modeling principles that are familiar from the explanation of animal and human intelligence. As these methods might serve as best practices for trustworthy AI, they deserve closer philosophical scrutiny.

Many of the computing systems programmed using Machine Learning are opaque: it is difficult to know why they do what they do or how they work. Explainable Artificial Intelligence aims to develop analytic techniques that render opaque computing systems transparent, but lacks a normative framework with which to evaluate these techniques' explanatory successes. The aim of the present discussion is to develop such a framework, paying particular attention to different stakeholders' distinct explanatory requirements. Building on an analysis of 'opacity' from philosophy of science, this framework is modeled after accounts of explanation in cognitive science. The framework distinguishes between the explanation-seeking questions that are likely to be asked by different stakeholders, and specifies the general ways in which these questions should be answered so as to allow these stakeholders to perform their roles in the Machine Learning ecosystem. By applying the normative framework to recently developed techniques such as input heatmapping, feature-detector visualization, and diagnostic classification, it is possible to determine whether and to what extent techniques from Explainable Artificial Intelligence can be used to render opaque computing systems transparent and, thus, whether they can be used to solve the Black Box Problem., Author(s): Carlos Zednik [sup.1] Author Affiliations: (1) grid.5807.a, 0000 0001 1018 4307, Otto-von-Guericke-Universität Magdeburg, , Magdeburg, Germany Introduction Computing systems programmed using Machine Learning (ML) are increasingly capable of solving [...]

The present article deals with the topic of Artificial Intelligence (AI) and its application areas. Both the societal opportunities and the risks of the development and use of artificially intelligent machines are mentioned. The use of AI has implications for various areas such as the economy, industry, the workplace, private life, and education. Political debates arise about the development and use of AI systems as well as the role of AI in future societies. The application of AI offers great opportunities but also carries societal risks. It is important for political decision-makers to regulate these technologies and, if necessary, prohibit them in order to prevent abuse and violation of rights. There are also debates about the sustainability of AI and questions about the nature of AI systems and the associated rights and obligations. It is important to create societal structures and institutions that regulate the handling of AI systems and consider both the rights and obligations of humans and the potential rights and obligations of AI systems. [Extracted from the article]

This paper considers the way mathematical and computational models are used in network neuroscience to deliver mechanistic explanations. Two case studies are considered: Recent work on klinotaxis by Caenorhabditis elegans, and a long-standing research effort on the network basis of schizophrenia in humans. These case studies illustrate the various ways in which network, simulation, and dynamical models contribute to the aim of representing and understanding network mechanisms in the brain, and thus, of delivering mechanistic explanations. After outlining this mechanistic construal of network neuroscience, two concerns are addressed. In response to the concern that functional network models are nonexplanatory, it is argued that functional network models are in fact explanatory mechanism sketches. In response to the concern that models which emphasize a network’s organization over its composition do not explain mechanistically, it is argued that this emphasis is both appropriate and consistent with the principles of mechanistic explanation. What emerges is an improved understanding of the ways in which mathematical and computational models are deployed in network neuroscience, as well as an improved conception of mechanistic explanation in general. [ABSTRACT FROM AUTHOR]

The article discusses how familiarity alters our perception of change. Topics discussed include division of change perception in three process modules including pre-processing, feature extraction, and identification stage, gating mechanism, processor interpretations of stage models, and the redirection of information from one stage to another.

The article "Artificial Intelligence" sheds light on the various aspects of Artificial Intelligence and its influence on the daily lives of people worldwide. It is pointed out that many people do not fully understand the concept of Artificial Intelligence, but are aware that it will have an immense impact on present and future life. The article examines the definitional, political-theoretical, and political-cultural aspects of Artificial Intelligence and raises questions about its significance, sociopolitical challenges, and war-ethical implications. It also discusses how the development and use of Artificial Intelligence can be responsibly promoted and regulated. [Extracted from the article]

The article comments on the article "Toward an experimental account of argumentation: the case of the slippery slope and the ad hominem arguments" by M. Lillo-Unglaube and others in the 2014 issue of the journal. Topics include the authors' stated to be arguing for the employment of more descriptive and experimental accounts on the psychology of argumentation, and the paper seeks to explain the rhetorical effectiveness of some of the informal fallacies.