Fabio Aurelio D'Asaro

Current research in Explainable AI includes post-hoc explanation methods that focus on building transparent explaining agents able to emulate opaque ones. Such agents are naturally required to be accurate and trustworthy. However, what it means for an explaining agent to be accurate and trustworthy is far from being clear. We characterize accuracy and trustworthiness as measures of the distance between the formal properties of a given opaque system and those of its transparent explanantes. To this aim, we extend Probabilistic Computation Tree Logic with operators to specify degrees of accuracy and trustworthiness of explaining agents. We also provide a semantics for this logic, based on a multi-agent structure and relative model-checking algorithms. The paper concludes with a simple example of a possible application.

A number of reasons, both historical and philosophical, has caused Technology and Human Sciences to be perceived as disjoint domains. Contrary to this claim, we propose as a thesis that there exists a strong methodological affinity between these apparently disconnected fields of knowledge. Our view is further corroborated by the emergence, in the setting of Information Sciences of new scientific concepts and tools such as fuzziness. Comparing the ways in which both technology and literature offer a “model of reality” we shall see that their approach is mainly descriptive, in opposition to the Hard Science’s approach that is markedly Galileian and proceeds by making bold hypotheses, not necessarily linked to the surface description of reality. Moreover, we will discuss the fact that, surprisingly, fuzzy logic can be applied in a very similar way in both Technology and Human Sciences – while its eventual embodiment in Hard Sciences is of a very different nature.