Bridging Cooperative Games, Team Optimization, and Machine Learning: Basic Concepts and Case Studies

• Machine Learning and Optimization: A fruitful interplay
• Cooperative Games: Main concepts and tools.
• Team Optimization: Main concepts and tools.
• Case studies.
  The choice of the case-studies will be based on the research interests of the PhD students following the Course. Here is a list of possible topics:
  o network congestion
  o multi-robot task allocation
  o reward assignment to interconnected autonomous vehicles
  o explainability in Reinforcement Learning
  o anomaly detection in explainable Artificial Intelligence
  o ensemble behavior in social groups
  o optimization-based scheduling
  o individual performance in group activities/sports
  o optimal production under uncertainty
  o resilience of complex networks
  o routing in data networks
  o movement analysis
  o non-verbal communication

A minimum of 75% attendance is required

Teaching methods

Lectures in presence. Class sessions involve a mix of methodological lectures and applications.

Assessment methods

Study of a a paper (a different paper assigned to each student) related to the topics of the Course and preparation of the material (20-30 slides) which illustrates critically its contents.

References

• Lecture notes of the instructor
• Papers on the case studies
• G. Owen: “Game Theory”. Emerald Group, 4th Edition, 2013
• R . Zoppoli, M. Sanguineti, G. Gnecco, T. Parisini: “Neural Approximations for Optimal Control and Decision”, Chapter 9: Team Optimal Control Problems. Springer, 2020