Agent Teams for Design Problems
University of Southern California / Viterbi School of Engineering
Leandro Soriano Marcolino, Haifeng Xu, Brian Kolev, Samori Price, Milind Tambe, Prof. David Gerber
Agent Based Modeling and Simulation, Generative Design, Office Tower Massing, Genetic Algorithms, Building Information Modelling
Description: Design imposes a novel social choice problem: using a team of voting agents, maximize the number of optimal solutions; allowing a user to then take an aesthetical choice. In an open system of design agents, team formation is fundamental. We present the first model of agent teams for design. For maximum applicability, we envision agents that are queried for a single opinion, and multiple solutions are obtained by multiple iterations.
We show that diverse teams composed of agents with different preferences maximize the number of optimal solutions, while uniform teams composed of multiple copies of the best agent are in general suboptimal. Our experiments study the model in bounded time; and we also study a real system, where agents vote to design buildings.