演算式決策與學習

As an emerging and active interdisciplinary research area, with contributions from theoretical computer science, economics, networking, artificial intelligence, operations research, and discrete mathematics, “algorithmic game theory” and “learning in multiagent systems” are focused on the analysis of equilibria such as efficiency of equilibria and complexity of computing equilibria, learning to reach equilibria in repeated games, or learning for design mechanism. In addition, we give a perspective on machine learning that treats “fairness” as a central concern. We will briefly review machine learning in a way that highlights ethical challenges, particularly, bias and even discrimination, with some approaches to mitigate these problems.

Evaluation and Grading Policy: Homework: 4 assignments (60%) Final Presentation: reading and presentation (40%)

• Introduction: Algorithmic decision
• Price of anarchy
• Computing equilibria & Learning in multiagent systems
• Multiagent systems
• Fairness and bias in machine learning
• Final presentation

Algorithmic Game Theory, edited by Noam Nisan, Tim Roughgarden, and Vijay V. Vazirani. 2007 Fairness and Machine Learning, by Solon Barocas, Moritz Hardt, and Arvind Narayanan. 2021 Handbook of Computational Social Choice. 2016 Multiagent Systems: Algorithmic, Game-Theoretic, and Logical Foundations, by Yoav Shohan and Kevin Leyton–Brown. 2009 References: Conference papers mainly from ACM EC, WINE, AAMAS, SAGT, STOC, FOCS, SODA, AAAI, etc. Journal papers mainly from GEB, IJGT, ACM TEAC, AIJ, JAIR, etc.

地點

TBD

時間

By appointment

聯絡方式

poanch@gmail.com