A Learning Framework for Distribution-Based Game-Theoretic Solution Concepts
This work provides a foundational methodology for learning economic solutions, which could impact researchers in machine learning and economics, though it appears incremental by generalizing existing tools.
The authors tackled the problem of learning economic solution concepts from data by developing a unified learning-theoretic framework, and they established conditions for PAC learnability, applying it to derive novel concepts like PAC competitive equilibrium and PAC Condorcet winners.
The past few years have seen several works on learning economic solutions from data; these include optimal auction design, function optimization, stable payoffs in cooperative games and more. In this work, we provide a unified learning-theoretic methodology for modeling such problems, and establish tools for determining whether a given economic solution concept can be learned from data. Our learning theoretic framework generalizes a notion of function space dimension -- the graph dimension -- adapting it to the solution concept learning domain. We identify sufficient conditions for the PAC learnability of solution concepts, and show that results in existing works can be immediately derived using our methodology. Finally, we apply our methods in other economic domains, yielding a novel notion of PAC competitive equilibrium and PAC Condorcet winners.