Manipulation-Proof Machine Learning

04/08/2020
by   Daniel Björkegren, et al.
0

An increasing number of decisions are guided by machine learning algorithms. In many settings, from consumer credit to criminal justice, those decisions are made by applying an estimator to data on an individual's observed behavior. But when consequential decisions are encoded in rules, individuals may strategically alter their behavior to achieve desired outcomes. This paper develops a new class of estimator that is stable under manipulation, even when the decision rule is fully transparent. We explicitly model the costs of manipulating different behaviors, and identify decision rules that are stable in equilibrium. Through a large field experiment in Kenya, we show that decision rules estimated with our strategy-robust method outperform those based on standard supervised learning approaches.

READ FULL TEXT
research
02/15/2017

Simple rules for complex decisions

From doctors diagnosing patients to judges setting bail, experts often b...
research
07/29/2020

Computing Optimal Decision Sets with SAT

As machine learning is increasingly used to help make decisions, there i...
research
12/28/2022

The Right to be an Exception to a Data-Driven Rule

Data-driven tools are increasingly used to make consequential decisions....
research
06/09/2017

Decoupling Learning Rules from Representations

In the artificial intelligence field, learning often corresponds to chan...
research
04/26/2021

Algorithm is Experiment: Machine Learning, Market Design, and Policy Eligibility Rules

Algorithms produce a growing portion of decisions and recommendations bo...
research
09/21/2020

Selectivity correction with online machine learning

Computer systems are full of heuristic rules which drive the decisions t...
research
08/20/2022

Rationalizing Pre-Analysis Plans: Statistical Decisions Subject to Implementability

Pre-analysis plans (PAPs) are a potential remedy to the publication of s...

Please sign up or login with your details

Forgot password? Click here to reset