All papers

Proceedings of the National Academy of Sciences

Bayesian modeling of human–AI complementarity

2022-03-11101 citationsDOI 10.1073/pnas.2111547119

Mark Steyvers · Heliodoro Tejeda · Gavin Kerrigan · Padhraic Smyth

Research summary

A Bayesian framework models when human and machine classifiers can complement one another while accounting for differences in expressed confidence. The result is useful for hybrid design but is centered on classification tasks.

Evidence strengths

  • Explicit human-machine comparison
  • Bayesian modeling of confidence
  • Reproducible code and data

Limitations

  • Centered on classification tasks
  • Modeling assumptions may not transfer to open-ended collaboration

Topics

  • Human-AI Interaction
  • Artificial Intelligence
  • Prediction
  • Cognitive Science

Connected claims

No public claims are connected to this paper yet.

Open publisher source