Abstract:Despite its important applications in Machine Learning, min-max optimization of nonconvex-nonconcave objectives remains elusive. Not only are there no known first-order methods converging even to approximate local min-max points, but the computational complexity of identifying them is also poorly understood. In this paper, we provide a characterization of the computational complexity of the problem, as well as of the limitations of first-order methods in constrained min-max optimization problems with nonconvex-nonconcave objectives and linear constraints. As a warm-up, we show that, even when the objective is a Lipschitz and smooth differentiable function, deciding whether a min-max point exists, in fact even deciding whether an approximate min-max point exists, is NP-hard. More importantly, we show that an approximate local min-max point of large enough approximation is guaranteed to exist, but finding one such point is PPAD-complete. The same is true of computing an approximate fixed point of Gradient Descent/Ascent. An important byproduct of our proof is to establish an unconditional hardness result in the Nemirovsky-Yudin model. We show that, given oracle access to some function $f : P \to [-1, 1]$ and its gradient $\nabla f$, where $P \subseteq [0, 1]^d$ is a known convex polytope, every algorithm that finds a $\varepsilon$-approximate local min-max point needs to make a number of queries that is exponential in at least one of $1/\varepsilon$, $L$, $G$, or $d$, where $L$ and $G$ are respectively the smoothness and Lipschitzness of $f$ and $d$ is the dimension. This comes in sharp contrast to minimization problems, where finding approximate local minima in the same setting can be done with Projected Gradient Descent using $O(L/\varepsilon)$ many queries. Our result is the first to show an exponential separation between these two fundamental optimization problems.

The Minimizer of the Sum of Two Strongly Convex Functions

The minimizer of the sum of two strongly convex functions

On the Set of Possible Minimizers of a Sum of Convex Functions

A Gradient Complexity Analysis for Minimizing the Sum of Strongly Convex Functions with Varying Condition Numbers

Greedy Minimization of Weakly Supermodular Set Functions

Incremental Subgradient Methods for Minimizing The Sum of Quasi-convex Functions

Near Optimal Stochastic Algorithms for Finite-Sum Unbalanced Convex-Concave Minimax Optimization

Minimizing Sum of Truncated Convex Functions and Its Applications

On Optimal Universal First-Order Methods for Minimizing Heterogeneous Sums

Tight Lower Complexity Bounds for Strongly Convex Finite-Sum Optimization

Convergence of alternating direction method for minimizing sum of two nonconvex functions with linear constraints

A Minimization Approach for Minimax Optimization with Coupled Constraints

Weakly-Convex Concave Min-Max Optimization: Provable Algorithms and Applications in Machine Learning

Efficient Algorithms for Sum-of-Minimum Optimization

A non-convex relaxed version of minimax theorems

Approximate Decomposable Submodular Function Minimization for Cardinality-Based Components

A Convexification-based Outer-Approximation Method for Convex and Nonconvex MINLP

Improved Algorithms for Convex-Concave Minimax Optimization

The Complexity of Constrained Min-Max Optimization

On the oracle complexity of smooth strongly convex minimization

Variational Theory and Algorithms for a Class of Asymptotically Approachable Nonconvex Problems