%0 Conference Proceedings
%T Stop Ordering Machine Learning Algorithms by Their Explainability! An Empirical Investigation of the Tradeoff Between Performance and Explainability
%+ Julius-Maximilians-Universität Würzburg (JMU)
%+ Otto-von-Guericke-Universität Magdeburg = Otto-von-Guericke University [Magdeburg] (OVGU)
%A Wanner, Jonas
%A Herm, Lukas-Valentin
%A Heinrich, Kai
%A Janiesch, Christian
%Z Part 4: Privacy and Transparency in a Digitised Society
%< avec comité de lecture
%( Lecture Notes in Computer Science
%B 20th Conference on e-Business, e-Services and e-Society (I3E)
%C Galway, Ireland
%Y Denis Dennehy
%Y Anastasia Griva
%Y Nancy Pouloudi
%Y Yogesh K. Dwivedi
%Y Ilias Pappas
%Y Matti Mäntymäki
%I Springer International Publishing
%3 Responsible AI and Analytics for an Ethical and Inclusive Digitized Society
%V LNCS-12896
%P 245-258
%8 2021-09-01
%D 2021
%R 10.1007/978-3-030-85447-8_22
%K Machine learning
%K Explainability
%K Performance
%K Tradeoff
%K User Study
%Z Computer Science [cs]
%Z Computer Science [cs]/Networking and Internet Architecture [cs.NI]Conference papers
%X Numerous machine learning algorithms have been developed and applied in the field. Their application indicates that there seems to be a tradeoff between their model performance and explainability. That is, machine learning models with higher performance are often based on more complex algorithms and therefore lack interpretability or explainability and vice versa. The true extent of this tradeoff remains unclear while some theoretical assumptions exist. With our research, we aim to explore this gap empirically with a user study. Using four distinct datasets, we measured the tradeoff for five common machine learning algorithms. Our two-factor factorial design considers low-stake and high-stake applications as well as classification and regression problems. Our results differ from the widespread linear assumption and indicate that the tradeoff between model performance and model explainability is much less gradual when considering end user perception. Further, we found it to be situational. Hence, theory-based recommendations cannot be generalized across applications.
%G English
%Z TC 6
%Z WG 6.11
%2 https://inria.hal.science/hal-03648118/document
%2 https://inria.hal.science/hal-03648118/file/512902_1_En_22_Chapter.pdf
%L hal-03648118
%U https://inria.hal.science/hal-03648118
%~ IFIP-LNCS
%~ IFIP
%~ IFIP-TC
%~ IFIP-WG
%~ IFIP-TC6
%~ IFIP-WG6-11
%~ IFIP-I3E
%~ IFIP-LNCS-12896