Too Many Specialists: Emergent Inefficiencies and Bottlenecks for Multi-agent Ad-hoc Collaboration
For researchers in multi-agent systems and collaboration, this work reveals emergent inefficiencies in ad-hoc teams, but the findings are incremental as they extend known concepts to a specific simulated domain.
The study uses an agent-based model of ad-hoc teamwork in a kitchen environment to show that heterogeneous agent traits and task structures create systemic bottlenecks, inefficiencies, and inequality, identifying a 'specialist's dilemma' where rigid role assertion worsens these issues.
Computational models of collaboration without prior coordination often overlook how heterogeneous agent traits and complex task structures jointly produce systemic bottlenecks, inefficiencies, and contribution inequalities. We address this by using an agent-based model of ad-hoc teamwork in a kitchen environment. Our model integrates diverse agent personas with tasks that combine serial and parallel dependencies. We identify a specialist's dilemma, where rigid role assertion generates system-level bottlenecks, amplifies workload inequality, and fosters fragmented, homophilous networks. We also find that team size and communication overhead interact with problem structure to generate diminishing returns and redundant collaboration. Linking micro-level behavior to macro-level outcomes provides insights into emergent collaboration and design principles for effective multi-agent teamwork.