Towards equity in algorithmic decision making
Abstract/Contents
- Abstract
- In recent years, many high-stakes societal decision-making systems have begun incorporating data and algorithms. This trend raises the question of how decision makers can do so in a way which creates equitable systems which ameliorate inequities. This dissertation considers two broad paths forward towards this goal. First, we review a series of interventions at various stages of the model-building and deployment process. Specifically, we consider how a model-builder might selectively acquire additional information, adaptively sample training data, and add personalization. We show that these interventions allow for model-builders to efficiently allocate resources to create decision-making systems which are inclusive of individuals from vulnerable groups. Second, we review two pieces of work where modern, online data sources give insights which can inform improvements for existing systems. In particular, we first consider how telematics data, containing records on the true prevalence of speeding, sheds light on inequities in traffic enforcement. Then, we see how online game records provide valuable insight into how users make decisions within social networks. Findings from both studies can be incorporated in future design of or interventions in decision-making systems within both spaces. Overall, this dissertation demonstrates two concrete paths for moving towards equitable decision making: intervening to efficiently improve outcomes for underserved groups, and leveraging insights from modern data sources to improve societal decision making systems.
Description
| Type of resource | text |
|---|---|
| Form | electronic resource; remote; computer; online resource |
| Extent | 1 online resource. |
| Place | California |
| Place | [Stanford, California] |
| Publisher | [Stanford University] |
| Copyright date | 2022; ©2022 |
| Publication date | 2022; 2022 |
| Issuance | monographic |
| Language | English |
Creators/Contributors
| Author | Cai, William |
|---|---|
| Degree supervisor | Goel, Sharad, 1977- |
| Degree supervisor | Ugander, Johan |
| Thesis advisor | Goel, Sharad, 1977- |
| Thesis advisor | Ugander, Johan |
| Thesis advisor | Ashlagi, Itai |
| Degree committee member | Ashlagi, Itai |
| Associated with | Stanford University, Department of Management Science and Engineering |
Subjects
| Genre | Theses |
|---|---|
| Genre | Text |
Bibliographic information
| Statement of responsibility | William Cai. |
|---|---|
| Note | Submitted to the Department of Management Science and Engineering. |
| Thesis | Thesis Ph.D. Stanford University 2022. |
| Location | https://purl.stanford.edu/tc218yn3796 |
Access conditions
- Copyright
- © 2022 by William Cai
- License
- This work is licensed under a Creative Commons Attribution Non Commercial 3.0 Unported license (CC BY-NC).
Also listed in
Loading usage metrics...