Research Projects

The late 20th century saw a shift in algorithmic domains, with researchers no longer dealing with optimization problems from heavy engineering projects or combinatorial modeling. The Internet economy, which generated revenue through keyword auctions, generated new types of problems involving multiple parties with competing objectives. Game Theory, traditionally focused on economic markets and autonomous agent interactions, was primarily the domain of economists, statisticians, and applied mathematicians. However, the large volume of interactions and fast-paced nature of the Internet market led to the need for fast and scalable computing. Theoretical Computer Science emerged as a new guiding light, focusing on average case analysis with worst case guarantees and appropriate trade-offs. Economics and Computation is a product of this synthesis, applying algorithm design and complexity theory principles to problems arising in Game Theory. In the last two decades, approximation algorithms have driven the growth and expansion of Algorithmic Game Theory. Applying Parameterized Complexity to Game Theory could further enhance and enrich the problem and computability of desirable solutions. This is particularly true for algorithms dealing with multi-agent systems, which must encode the vast number of choices of agents and the payoff scenarios. Parameterization of secondary and tertiary aspects of the problem, beyond input size, can provide additional information about structure or value from existing trends, economic forecasts, or historical data. This project aims to hold a multivariate computational lens to Economics and Computation, focusing on problems modeled as matching, auction, and contracts, the central instruments of the global economy.