Date of Degree

5-2019

Document Type

Dissertation

Degree Name

Ph.D.

Program

Computer Science

Advisor

Simon Parsons

Committee Members

Noson Yanofsky

Olympia Hadjiliadis

Gennaro Rosario

Peter McBurney

Subject Categories

Computational Engineering | Social and Behavioral Sciences

Keywords

Mechanism Design, Transfer Learning, Network Market

Abstract

Mechanism design is the sub-field of microeconomics and game theory, which considers agents have their own private information and are self-interested and tries to design systems that can produce desirable outcomes. In recent years, with the development of internet and electronic markets, mechanism design has become an important research field in computer science. This work has largely focused on single markets. In the real world, individual markets tend to connect to other markets and form a big “network market”, where each market occupies a node in the network and connections between markets reflect constraints on traders in the markets. So, it is interesting to find out how the structure of connected network markets impacts the performance of the resulting network markets and how we can optimize performance by varying the things that one could control in a network market. In this dissertation, I aim to find out whether we can apply transfer learning to other machine learning techniques like reinforcement learning in the design of network markets to help optimize the performance of the network markets. I applied transfer learning on both machine learning trading strategies and machine learning strategies for selecting which market to trade in. I found that, in most cases, by applying transfer learning to machine learning trading strategies or machine learning market selection strategies, we can improve the performance of the network market significantly.

Share

COinS