pySOT and POAP: An event-driven asynchronous framework for surrogate optimization

TL;DR

This paper introduces POAP, an event-driven asynchronous framework, and pySOT, a toolbox for surrogate optimization, enabling efficient global optimization of expensive functions with parallel and asynchronous evaluations.

Contribution

The paper presents a novel asynchronous framework (POAP) and a comprehensive surrogate optimization toolbox (pySOT) supporting various models and strategies for parallel global optimization.

Findings

Asynchronous computation advantage increases with evaluation time variance and processor count.

Close to linear speed-up observed with 4, 8, and 16 processors.

Extensive comparison between synchronous and asynchronous parallelism.

Abstract

This paper describes Plumbing for Optimization with Asynchronous Parallelism (POAP) and the Python Surrogate Optimization Toolbox (pySOT). POAP is an event-driven framework for building and combining asynchronous optimization strategies, designed for global optimization of expensive functions where concurrent function evaluations are useful. POAP consists of three components: a worker pool capable of function evaluations, strategies to propose evaluations or other actions, and a controller that mediates the interaction between the workers and strategies. pySOT is a collection of synchronous and asynchronous surrogate optimization strategies, implemented in the POAP framework. We support the stochastic RBF method by Regis and Shoemaker along with various extensions of this method, and a general surrogate optimization strategy that covers most Bayesian optimization methods. We have implemented many different surrogate models, experimental designs, acquisition functions, and a large set of test problems. We make an extensive comparison between synchronous and asynchronous parallelism and find that the advantage of asynchronous computation increases as the variance of the evaluation time or number of processors increases. We observe a close to linear speed-up with 4, 8, and 16 processors in both the synchronous and asynchronous setting.