Game Theory With Apache Spark, Part 1
Posted on November 9, 2018
In general terms, Game Theory is concerned with decision making in an environment that involves multiple entities with possibly conflicting interests (Myerson, Roger B. (1997). Game Theory: Analysis of Conflict, Harvard University Press, Wikipedia). Game theory is particularly useful in Prescriptive Analytics where techniques from game theory can be employed to find optimal solutions to complicated decision making problems (for example, see this discussion).
In this article, we will formulate a general resource allocation problem that can be solved using a particular game theory algorithm based on economist Ausubel’s Efficient Dynamic Auction Method. We will then implement the algorithm in Apache Spark Machine Learning Library using Java and demonstrate how it works via examples.