Transparency in Complex Computational Systems
Published online by Cambridge University Press: 01 January 2022
Scientists depend on complex computational systems that are often ineliminably opaque, to the detriment of our ability to give scientific explanations and detect artifacts. Some philosophers have suggested treating opaque systems instrumentally, but computer scientists developing strategies for increasing transparency are correct in finding this unsatisfying. Instead, I propose an analysis of transparency as having three forms: transparency of the algorithm, the realization of the algorithm in code, and the way that code is run on particular hardware and data. This targets the transparency most useful for a task, avoiding instrumentalism by providing partial transparency when full transparency is impossible.
- Research Article
- Copyright © The Philosophy of Science Association
I am grateful for helpful comments from and discussions with Holly Andersen, Robert Batterman, Nora Mills Boyd, Liam Kofi Bright, Mazviita Chirimuuta, Roger Creel, Javier Duarte, Mahi Hardalupas, Paul Humphreys, Benjamin Jantzen, Johannes Lenhard, Sabina Leonelli, Jake Levinson, Edouard Machery, Sandra Mitchell, Elinor Nichols, Kathleen Nichols, Aaron Novick, Olivia Ordoñez, William Penn, Rebecca Traber, Porter Williams, Eric Winsberg, and two anonymous reviewers. Thanks also to generous audiences at Philosophical Perspectives on Data-Intensive Science in Hannover; Models and Simulations 8 in Columbia, SC; the Machine Learning Workshop in Irvine, CA; and Science and Art of Simulation IV in Stuttgart.