A QUBO Framework for Team Formation

2025-03-29

AI Summary

This paper addresses the skill-driven team formation problem: given a set of experts—each possessing a distinct skill set—and a task requiring specific skills, the goal is to select a cost-optimal subset of experts that maximizes skill coverage. To overcome the fragmentation in existing works caused by disparate cost definitions, we propose the first unified Quadratic Unconstrained Binary Optimization (QUBO) modeling framework, accommodating three representative cost models—e.g., cardinality (team size), total monetary cost, and heterogeneous expert-specific costs. Our key innovation lies in embedding multiple cost formulations uniformly into the QUBO objective and integrating a Graph Neural Network (GNN) to learn structured expert-skill representations, enabling cross-task transfer learning. Experiments demonstrate that our QUBO-based solver achieves performance on par with conventional baselines; moreover, the GNN-enhanced variant significantly improves efficiency on unseen tasks, validating the framework’s scalability and generalization capability.

Technology Category

• Humans and AI: Teamwork, Team formation
• Search and Optimization: Learning to Search
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Economics, Online Markets and Human Computation: Trust and reliance of crowd workers and data experts on GenAI
• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies

Abstract

The team formation problem assumes a set of experts and a task, where each expert has a set of skills and the task requires some skills. The objective is to find a set of experts that maximizes coverage of the required skills while simultaneously minimizing the costs associated with the experts. Different definitions of cost have traditionally led to distinct problem formulations and algorithmic solutions. We introduce the unified TeamFormation formulation that captures all cost definitions for team formation problems that balance task coverage and expert cost. Specifically, we formulate three TeamFormation variants with different cost functions using quadratic unconstrained binary optimization (QUBO), and we evaluate two distinct general-purpose solution methods. We show that solutions based on the QUBO formulations of TeamFormation problems are at least as good as those produced by established baselines. Furthermore, we show that QUBO-based solutions leveraging graph neural networks can effectively learn representations of experts and skills to enable transfer learning, allowing node embeddings from one problem instance to be efficiently applied to another.

Problem

• Unified framework for team formation balancing skills and costs
• QUBO formulations for three TeamFormation variants with different costs
• Graph neural networks enable transfer learning in team formation

Innovation

• Unified QUBO framework for team formation
• Three variants with different cost functions
• Graph neural networks enable transfer learning