상세정보
Export to Refworks부가기능
Fairness and Feedback in Learning and Games
상세 프로파일
| 자료유형 | 학위논문 |
|---|---|
| 서명/저자사항 | Fairness and Feedback in Learning and Games. |
| 개인저자 | Jabbari, Shahin. |
| 단체저자명 | University of Pennsylvania. Computer and Information Science. |
| 발행사항 | [S.l.]: University of Pennsylvania., 2019. |
| 발행사항 | Ann Arbor: ProQuest Dissertations & Theses, 2019. |
| 형태사항 | 117 p. |
| 기본자료 저록 | Dissertations Abstracts International 81-05B. Dissertation Abstract International |
| ISBN | 9781088355619 |
| 학위논문주기 | Thesis (Ph.D.)--University of Pennsylvania, 2019. |
| 일반주기 |
Source: Dissertations Abstracts International, Volume: 81-05, Section: B.
Advisor: Kearns, Michael. |
| 이용제한사항 | This item must not be sold to any third party vendors. |
| 요약 | In this thesis, we study fairness and feedback effects in game theory and machine learning. In game theory and economics, financial or technological networks are analyzed for feedback effects. These studies analyze how the connectivity benefits or risk of contagious shocks affect the individual agents or the structure of the network formed by these rational agents. Towards this direction, in the first part of this thesis, we study a series of novel network formation games and analyze the structural properties of the equilibrium networks.Feedback effects can also occur in machine learning problems such as reinforcement learning or sequential allocation problems where the decisions of an algorithm over time can change the resources or actions available to the algorithm in the future as well as the environment in which the algorithm is operating. In the second part of this thesis, we study the effect of these feedback loops and ways to prevent them while also ensuring that the algorithm's actions and allocations satisfy natural notions of fairness. In particular we are interested in quantifying the cost of imposing fairness on learning algorithms. |
| 일반주제명 | Computer science. |
| 언어 | 영어 |
| 바로가기 | 
: 이 자료의 원문은 한국교육학술정보원에서 제공합니다. |
