OpenBox: A Generalized Black-box Optimization Service

TL;DR

OpenBox is a scalable, fault-tolerant, open-source black-box optimization service that enhances usability and efficiency through modular design, parallelization, and transfer learning, outperforming existing systems.

Contribution

The paper introduces OpenBox, a versatile BBO platform with a modular architecture, distributed scalability, and novel efficiency techniques like algorithm-agnostic parallelization and transfer learning.

Findings

OpenBox outperforms existing BBO systems in effectiveness.

OpenBox demonstrates improved efficiency through parallelization and transfer learning.

OpenBox is scalable and fault-tolerant, suitable for diverse applications.

Abstract

Black-box optimization (BBO) has a broad range of applications, including automatic machine learning, engineering, physics, and experimental design. However, it remains a challenge for users to apply BBO methods to their problems at hand with existing software packages, in terms of applicability, performance, and efficiency. In this paper, we build OpenBox, an open-source and general-purpose BBO service with improved usability. The modular design behind OpenBox also facilitates flexible abstraction and optimization of basic BBO components that are common in other existing systems. OpenBox is distributed, fault-tolerant, and scalable. To improve efficiency, OpenBox further utilizes "algorithm agnostic" parallelization and transfer learning. Our experimental results demonstrate the effectiveness and efficiency of OpenBox compared to existing systems.