Search papers, labs, and topics across Lattice.
This paper establishes that finding approximate stationary points in min-max optimization problems involving quadratic polynomials is PPAD-hard, even under specific constraints such as multilinearity and limited variable involvement. This result is significant because it extends the complexity landscape of optimization problems, particularly in the context of game theory. Additionally, it provides the first PPAD-hardness results for two-team zero-sum polymatrix games, highlighting the inherent difficulty of these scenarios.
Computing approximate stationary points in min-max optimization for quadratic polynomials is proven to be PPAD-hard, revealing deep implications for game theory and optimization.
We prove that computing approximate stationary points of min-max optimization over the hypercube is PPAD-hard for quadratic polynomials. This holds even when the polynomials are multilinear, each variable appears in at most three monomials, and the approximation factor is inverse polynomial. As a direct consequence, we obtain the first PPAD-hardness results for two-team zero-sum polymatrix games.
