GPU-Accelerated SPOCK for Scenario-Based Risk-Averse Optimal Control Problems

TL;DR

This paper introduces a GPU-accelerated version of the SPOCK algorithm, significantly improving the speed and memory efficiency for large-scale scenario-based risk-averse optimal control problems through parallelization.

Contribution

It presents a novel GPU implementation of SPOCK, demonstrating enhanced performance and analyzing the impact of scenario tree structures on parallelization efficiency.

Findings

GPU-accelerated SPOCK outperforms traditional solvers in speed

Memory footprint is reduced with GPU implementation

Scenario tree structure affects parallelization and solve time

Abstract

This paper presents a GPU-accelerated implementation of the SPOCK algorithm, a proximal method designed for solving scenario-based risk-averse optimal control problems. The proposed implementation leverages the massive parallelization of the SPOCK algorithm, and benchmarking against state-of-the-art interior-point solvers demonstrates GPU-accelerated SPOCK's competitive execution time and memory footprint for large-scale problems. We further investigate the effect of the scenario tree structure on parallelizability, and so on solve time.