Algorithmic encounters: An interactional approach to the AI accuracy vs interpretability trade-off

Gian Marco Campagnolo, Edward Sharkey

Research output: Working paper

Abstract

In this paper we attempt to provide a formal assessment of AI interpretability based on an observation of machine learning development sessions as 'algorithmic encounters'. We begin by considering the unbalance between the existing lacuna of metrics to assess AI interpretability in relation to the abundance of established approaches for measuring accuracy. To this respect, we introduce an approach based on interactional sociology to the study of AI interpretability as the patterning of talk between different forms of machine learning models utilisation: reproductive and consumptive utilisation. We then provide a detailed scenario of how this approach could be designed into a formal analysis of AI interpretability, based on the example of the testing of rival models to predict football player transfer value.
Original languageEnglish
Publication statusIn preparation - 2021

Keywords

  • accuracy
  • interpretability
  • symbolic interactionism
  • artificial intelligence
  • gaming encounters

Fingerprint

Dive into the research topics of 'Algorithmic encounters: An interactional approach to the AI accuracy vs interpretability trade-off'. Together they form a unique fingerprint.

Cite this