Philosophy and Technology 34 (2):265-288 (2021)
AbstractMany 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.
Similar books and articles
Solving the Black Box Problem: A Normative Framework for Explainable Artificial Intelligence.Carlos Zednik - 2019 - Philosophy and Technology 34 (2):265-288.
Black-box artificial intelligence: an epistemological and critical analysis.Manuel Carabantes - 2020 - AI and Society 35 (2):309-317.
AISC 17 Talk: The Explanatory Problems of Deep Learning in Artificial Intelligence and Computational Cognitive Science: Two Possible Research Agendas.Antonio Lieto - 2018 - In Proceedings of AISC 2017.
The Pragmatic Turn in Explainable Artificial Intelligence (XAI).Andrés Páez - 2019 - Minds and Machines 29 (3):441-459.
Philosophy of Artificial Intelligence: A Critique of the Mechanistic Theory of Mind.Rajakishore Nath - 2009 - Universal Publishers.
A framework for the foundation of the philosophy of artificial intelligence.Viola Schiaffonati - 2003 - Minds and Machines 13 (4):537-552.
Moral orthoses: A new approach to human and machine ethics.Marius Dorobantu & Yorick Wilks - 2019 - Zygon 54 (4):1004-1021.
Universal intelligence: A definition of machine intelligence.Shane Legg & Marcus Hutter - 2007 - Minds and Machines 17 (4):391-444.
The Facets of Artificial Intelligence: A Framework to Track the Evolution of AI.Fernando Martínez-Plumed, Bao Sheng Loe, Peter Flach, Sean O. O. HEigeartaigh, Karina Vold & José Hernández-Orallo - 2018 - In Proceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence Evolution of the contours of AI. pp. 5180-5187.
A competence framework for artificial intelligence research.Lisa Miracchi - 2019 - Philosophical Psychology 32 (5):588-633.
Philosophy and theory of artificial intelligence 2017.Vincent C. Müller (ed.) - 2017 - Berlin: Springer.
Added to PP
Historical graph of downloads
Citations of this work
Artificial intelligence, transparency, and public decision-making.Karl de Fine Licht & Jenny de Fine Licht - 2020 - AI and Society 35 (4):917-926.
What is Interpretability?Adrian Erasmus, Tyler D. P. Brunet & Eyal Fisher - 2021 - Philosophy and Technology 34:833–862.
Transparency and the Black Box Problem: Why We Do Not Trust AI.Warren J. von Eschenbach - 2021 - Philosophy and Technology 34 (4):1607-1622.
Defining the undefinable: the black box problem in healthcare artificial intelligence.Jordan Joseph Wadden - 2022 - Journal of Medical Ethics 48 (10):764-768.
Scientific Exploration and Explainable Artificial Intelligence.Carlos Zednik & Hannes Boelsen - 2022 - Minds and Machines 32 (1):219-239.
References found in this work
Eliminative Materialism and Propositional Attitudes.Paul M. Churchland - 1981 - Journal of Philosophy 78 (2):67-90.
On the proper treatment of connectionism.Paul Smolensky - 1988 - Behavioral and Brain Sciences 11 (1):1-23.
How the machine ‘thinks’: Understanding opacity in machine learning algorithms.Jenna Burrell - 2016 - Big Data and Society 3 (1):205395171562251.
Integrating psychology and neuroscience: functional analyses as mechanism sketches.Gualtiero Piccinini & Carl Craver - 2011 - Synthese 183 (3):283-311.