基于强化学习的可扩展神经架构搜索
Scalable Reinforcement Learning-based Neural Architecture Search
AI总结:
本文提出一种基于强化学习的神经架构搜索方法,智能体学习搜索良好架构而非返回单一最优解,实验表明该方法在搜索空间大小方面具有强可扩展性,但对超参数变化的鲁棒性有限。
摘要:
本文评估了一种新颖的基于强化学习的神经架构搜索(Neural Architecture Search, NAS)解决方案的能力。在该方案中,强化学习(Reinforcement Learning, RL)智能体学习搜索良好的架构,而非返回单一的最优架构。我们考虑了 NAS-Bench-101 和 NAS-Bench-301 两种设置,并与局部搜索和随机搜索等各种已知的强基线进行了比较。我们得出结论,我们的强化学习智能体在搜索空间大小方面表现出强大的可扩展性,但对超参数变化的鲁棒性有限。
英文摘要:
In this publication, we assess the ability of a novel Reinforcement Learning-based solution to the problem of Neural Architecture Search, where a Reinforcement Learning (RL) agent learns to search for good architectures, rather than to return a single optimal architecture. We consider both the NAS-Bench-101 and NAS- Bench-301 settings, and compare against various known strong baselines, such as local search and random search. We conclude that our Reinforcement Learning agent displays strong scalability with regards to the size of the search space, but limited robustness to hyperparameter changes.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。