arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2001.08437cs.LGstat.ML
收藏

基于非平稳策略梯度的多目标神经网络结构搜索

Multi-objective Neural Architecture Search via Non-stationary Policy Gradient

Huawei Noah’s Ark Lab(华为诺亚方舟实验室)

Zewei Chen, Fengwei Zhou, George Trimponias, Zhenguo Li

摘要:

多目标神经架构搜索(NAS)旨在发现存在多个冲突目标的新架构。尽管最近取得了进展,但准确有效地近似完整帕累托前沿的问题仍然具有挑战性。在这项工作中,我们探讨了新的强化学习(RL)的非平稳策略梯度(NPG)的基础上的范例。NPG利用一个非平稳的奖励函数,并鼓励不断适应的政策,以有效地捕捉整个帕累托前沿。我们引入了两个新的奖励函数的元素从标量化和进化的主导范式。为了处理非平稳性,我们提出了一个新的探索计划,使用余弦温度衰减与热重启。为了快速准确的架构评估,我们引入了一个新的预训练共享模型,我们在整个训练过程中不断进行微调。我们对各种数据集的广泛实验研究表明,我们的框架可以快速近似完整的Pareto前沿。此外,与其他多目标NAS方法和类似网络规模的其他单目标NAS方法相比,我们发现的细胞可以实现最高的预测性能。我们的工作证明了NPG作为一个简单,高效,有效的多目标NAS范例的潜力。

英文摘要:

Multi-objective Neural Architecture Search (NAS) aims to discover novel architectures in the presence of multiple conflicting objectives. Despite recent progress, the problem of approximating the full Pareto front accurately and efficiently remains challenging. In this work, we explore the novel reinforcement learning (RL) based paradigm of non-stationary policy gradient (NPG). NPG utilizes a non-stationary reward function, and encourages a continuous adaptation of the policy to capture the entire Pareto front efficiently. We introduce two novel reward functions with elements from the dominant paradigms of scalarization and evolution. To handle non-stationarity, we propose a new exploration scheme using cosine temperature decay with warm restarts. For fast and accurate architecture evaluation, we introduce a novel pre-trained shared model that we continuously fine-tune throughout training. Our extensive experimental study with various datasets shows that our framework can approximate the full Pareto front well at fast speeds. Moreover, our discovered cells can achieve supreme predictive performance compared to other multi-objective NAS methods, and other single-objective NAS methods at similar network sizes. Our work demonstrates the potential of NPG as a simple, efficient, and effective paradigm for multi-objective NAS.

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。

↑