Corpus ID: 232240535

Differentiable Learning Under Triage

@article{Okati2021DifferentiableLU,
  title={Differentiable Learning Under Triage},
  author={Nastaran Okati and A. De and M. Gomez-Rodriguez},
  journal={ArXiv},
  year={2021},
  volume={abs/2103.08902}
}
Multiple lines of evidence suggest that predictive models may benefit from algorithmic triage. Under algorithmic triage, a predictive model does not predict all instances but instead defers some of them to human experts. However, the interplay between the prediction accuracy of the model and the human experts under algorithmic triage is not well understood. In this work, we start by formally characterizing under which circumstances a predictive model may benefit from algorithmic triage. In… Expand

Figures from this paper

References

SHOWING 1-10 OF 37 REFERENCES
Regression Under Human Assistance
  • 7
  • Highly Influential
  • PDF
Classification Under Human Assistance
  • 1
  • PDF
The Algorithmic Automation Problem: Prediction, Triage, and Human Effort
  • 13
  • Highly Influential
  • PDF
Optimizing AI for Teamwork
  • 8
  • Highly Influential
  • PDF
Learning to Complement Humans
  • 14
  • PDF
Consistent Estimators for Learning to Defer to an Expert
  • 16
  • Highly Influential
  • PDF
Consistent algorithms for multiclass classification with an abstain option
  • 17
  • PDF
Ask Not What AI Can Do, But What AI Should Do: Towards a Framework of Task Delegability
  • 9
  • PDF
Active Regression by Stratification
  • 27
  • PDF
Deep Gamblers: Learning to Abstain with Portfolio Theory
  • 28
  • PDF
...
1
2
3
4
...