Consistent Estimators for Learning to Defer to an Expert

Learning algorithms are often used in conjunction with expert decision makers in practical scenarios, however, this fact is largely ignored when designing these algorithms. In this thesis, we explore how to learn predictors that can either predict or choose to defer the decision to a downstream expert. Given only samples of the expert's decisions, we give a procedure based on learning a classifier and a rejector and analyze it theoretically. Our approach is based on a novel reduction to cost sensitive learning where we give a consistent surrogate loss for cost sensitive learning that generalizes the cross entropy loss. We show the effectiveness of our approach on a variety of experimental tasks.

Name thesis:degree_name

Master

Department dc:contributor.department

Massachusetts Institute of Technology. Institute for Data, Systems, and Society

Grantor dc:publisher

Massachusetts Institute of Technology

Year dc:date.issued

2023

Author dc:creator

• Mozannar, Hussein

Advisor dc:contributor.advisor

• Sontag, David

Statement dc:rights

• Attribution-ShareAlike 4.0 International (CC BY-SA 4.0)
• Copyright retained by author(s)

Licence dc:rights.uri

• https://creativecommons.org/licenses/by-sa/4.0/

Handle dc:identifier.uri

https://hdl.handle.net/1721.1/151827

OAI identifier oai:identifier

oai:dspace.mit.edu:1721.1/151827