Stochastic Compositional Optimization with Compositional Constraints

TL;DR

This paper introduces a new stochastic compositional optimization framework with complex constraints, providing algorithms with convergence guarantees for risk management and data-driven applications.

Contribution

It extends SCO to handle expectation-based and compositional constraints, proposing primal-dual algorithms with proven convergence rates.

Findings

Algorithms achieve $ ext{O}(1/\sqrt{N})$ convergence rate.

Framework applicable to risk-averse optimization and portfolio selection.

Establishes benchmark results for constrained SCO.

Abstract

Stochastic compositional optimization (SCO) has attracted considerable attention because of its broad applicability to important real-world problems. However, existing works on SCO assume that the projection within a solution update is simple, which fails to hold for problem instances where the constraints are in the form of expectations, such as empirical conditional value-at-risk constraints. We study a novel model that incorporates single-level expected value and two-level compositional constraints into the current SCO framework. Our model can be applied widely to data-driven optimization and risk management, including risk-averse optimization and high-moment portfolio selection, and can handle multiple constraints. We further propose a class of primal-dual algorithms that generates sequences converging to the optimal solution at the rate of $\cO(\frac{1}{\sqrt{N}})$under both single-level expected value and two-level compositional constraints, where $N$ is the iteration counter, establishing the benchmarks in expected value constrained SCO.