Many evolutionary algorithms (EAs) take advantage of parallel evaluation of candidates. However, if evaluation times vary significantly, many worker nodes (i.e.,\ compute clients) are idle much of the time, waiting for the next generation to be created. Evolutionary neural architecture search (ENAS), a class of EAs that optimizes the architecture and hyperparameters of deep neural networks, is particularly vulnerable to this issue. This paper proposes a generic asynchronous evaluation strategy (AES) that is then adapted to work with ENAS. AES increases throughput by maintaining a queue of upto $K$ individuals ready to be sent to the workers for evaluation and proceeding to the next generation as soon as $M<<K$ individuals have been evaluated by the workers. A suitable value for $M$ is determined experimentally, balancing diversity and efficiency. To showcase the generality and power of AES, it was first evaluated in 11-bit multiplexer design (a single-population verifiable discovery task) and then scaled up to ENAS for image captioning (a multi-population open-ended-optimization task). In both problems, a multifold performance improvement was observed, suggesting that AES is a promising method for parallelizing the evolution of complex systems with long and variable evaluation times, such as those in ENAS.

翻译：许多进化算法（EAs）利用候选解的并行评估来提高效率。然而，若评估时间差异显著，大量工作节点（即计算客户端）常处于空闲状态，等待下一代个体生成。作为一类优化深度神经网络架构与超参数的进化算法，进化神经架构搜索（ENAS）对此问题尤为敏感。本文提出一种通用异步评估策略（AES），并将其适配至ENAS框架。该策略通过维护一个最多包含$K$个待评估个体的队列，并允许在工人节点完成$M<<K$个体评估后立即进入下一代，从而提升系统吞吐量。通过实验确定$M$的适宜取值以平衡多样性及效率。为验证AES的通用性与有效性，首先在11位多路复用器设计（单种群可验证发现任务）上评估，随后扩展至图像描述生成的ENAS（多种群开放式优化任务）。实验表明，在两类问题中均观察到多倍性能提升，表明AES是并行化评估时间较长且波动显著的复杂系统（如ENAS中所述场景）进化过程的有效方法。