用于等离子体科学表示学习的全卷积时空模型
Fully Convolutional Spatio-Temporal Models for Representation Learning in Plasma Science
摘要：
我们已经训练了一个完全卷积的时空模型，用于在聚变能等离子体科学这一具有挑战性的典型应用领域中进行快速准确的表示学习。大破坏的发生是先进托卡马克必须解决的一个至关重要的聚变能科学（FES）问题。虽然各种统计方法已被用于解决托卡马克中断预测和控制问题，但最近基于深度学习的方法已被证明特别引人注目。在本文中，我们介绍了融合递归神经网络（FRNN）软件套件的进一步改进。到目前为止，FRNN是基于循环神经网络的长短期记忆（LSTM）变体，以利用数据中的时间信息。在这里，我们实现并将时间卷积神经网络（TCN）架构应用于时间相关的输入信号，从而使FRNN架构完全卷积。这允许高度优化的卷积运算承载训练的大部分计算负载，从而能够减少训练时间，并有效使用高性能计算（HPC）资源进行超参数调整。与此同时，与LSTM架构相比，基于TCN的架构在大型代表性融合数据库中实现了相等或更好的预测性能。在数据丰富的科学学科中，这些结果对基于深度学习的一般时空特征提取器的资源有效训练具有影响。此外，这个具有挑战性的示例性案例研究说明了具有灵活架构选择选项的预测平台的优势，该平台能够轻松调整和适应，以响应大型现代观察数据集中日益出现的预测需求。
英文摘要：
We have trained a fully convolutional spatio-temporal model for fast and accurate representation learning in the challenging exemplar application area of fusion energy plasma science. The onset of major disruptions is a critically important fusion energy science (FES) issue that must be resolved for advanced tokamak. While a variety of statistical methods have been used to address the problem of tokamak disruption prediction and control, recent approaches based on deep learning have proven particularly compelling. In the present paper, we introduce further improvements to the fusion recurrent neural network (FRNN) software suite. Up to now, FRNN was based on the long short-term memory (LSTM) variant of recurrent neural networks to leverage the temporal information in the data. Here, we implement and apply the temporal convolutional neural network (TCN) architecture to the time-dependent input signals, thus rendering the FRNN architecture fully convolutional. This allows highly optimized convolution operations to carry the majority of the computational load of training, thus enabling a reduction in training time, and the effective use of high performance computing (HPC) resources for hyperparameter tuning. At the same time, the TCN based architecture achieves equal or better predictive performance when compared with the LSTM architecture for a large, representative fusion database. Across data-rich scientific disciplines, these results have implications for the resource-effective training of general spatio-temporal feature extractors based on deep learning. Moreover, this challenging exemplar case study illustrates the advantages of a predictive platform with flexible architecture selection options capable of being readily tuned and adapted for responding to prediction needs that increasingly arise in large modern observational dataset.
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。
