19 October 2020 in WIREs Data Mining and Knowledge Discovery
Explainability in Artificial Intelligence (AI) has been revived as a topic of active research by the need of conveying safety and trust to users in the “how” and “why” of automated decision‐making in different applications such as autonomous driving, medical diagnosis, or banking and finance. While explainability in AI has recently received significant attention, the origins of this line of work go back several decades to when AI systems were mainly developed as (knowledge‐based) expert systems. Since then, the definition, understanding, and implementation of explainability have been picked up in several lines of research work, namely, expert systems, machine learning, recommender systems, and in approaches to neural‐symbolic learning and reasoning, mostly happening during different periods of AI history. In this article, we present a historical perspective of Explainable Artificial Intelligence. We discuss how explainability was mainly conceived in the past, how it is understood in the present and, how it might be understood in the future. We conclude the article by proposing criteria for explanations that we believe will play a crucial role in the development of human‐understandable explainable systems.This article is categorized under: Fundamental Concepts of Data and Knowledge > Explainable AI Technologies > Artificial Intelligence
1 July 2021 in Artificial Intelligence
FMFermín Moscoso del Prado Martín
Fermín Moscoso del Prado Martín
7 May 2018 in Machine Learning
During the 1980s Michie defined Machine Learning in terms of two orthogonal axes of performance: predictive accuracy and comprehensibility of generated hypotheses. Since predictive accuracy was readily measurable and comprehensibility not so, later definitions in the 1990s, such as Mitchell’s, tended to use a one-dimensional approach to Machine Learning based solely on predictive accuracy, ultimately favouring statistical over symbolic Machine Learning approaches. In this paper we provide a definition of comprehensibility of hypotheses which can be estimated using human participant trials. We present two sets of experiments testing human comprehensibility of logic programs. In the first experiment we test human comprehensibility with and without predicate invention. Results indicate comprehensibility is affected not only by the complexity of the presented program but also by the existence of anonymous predicate symbols. In the second experiment we directly test whether any state-of-the-art ILP systems are ultra-strong learners in Michie’s sense, and select the Metagol system for use in humans trials. Results show participants were not able to learn the relational concept on their own from a set of examples but they were able to apply the relational definition provided by the ILP system correctly. This implies the existence of a class of relational concepts which are hard to acquire for humans, though easy to understand given an abstract explanation. We believe improved understanding of this class could have potential relevance to contexts involving human learning, teaching and verbal interaction.
2 March 2023 in IEEE Transactions on Information Forensics and Security
Despite the large body of academic work on machine learning security, little is known about the occurrence of attacks on machine learning systems in the wild. In this paper, we report on a quantitative study with 139 industrial practitioners. We analyze attack occurrence and concern and evaluate statistical hypotheses on factors influencing threat perception and exposure. Our results shed light on real-world attacks on deployed machine learning. On the organizational level, while we find no predictors for threat exposure in our sample, the amount of implement defenses depends on exposure to threats or expected likelihood to become a target. We also provide a detailed analysis of practitioners’ replies on the relevance of individual machine learning attacks, unveiling complex concerns like unreliable decision making, business information leakage, and bias introduction into models. Finally, we find that on the individual level, prior knowledge about machine learning security influences threat perception. Our work paves the way for more research about adversarial machine learning in practice, but yields also insights for regulation and auditing.
22 April 2015 in KI - Künstliche Intelligenz