RTML: Small: Collaborative: A Programming Model and Platform Architecture for Real-time Machine Learning for Sub-second Systems

RTML：小型：协作：亚秒级系统实时机器学习的编程模型和平台架构

• 批准号: 1937460
• 负责人: Simon Laflamme
• 金额: $ 24万
• 依托单位: Iowa State University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2019
• 资助国家: 美国
• 起止时间: 2019-10-01 至 2024-09-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1937460&HistoricalAwards=false
• 关键词: RTML; Small; Collaborative; Programming; Model

This project develops and evaluates novel frameworks for achieving real-time machine learning; that is, for a given target application that is producing a lot of data, how to process that data to concurrently prediction what comes next while learning from the past data at the same pace of the target application. The developed framework will produce adaptive models suitable to predict the behavior of the complex dynamics found in sub-second systems. Such systems include adaptive airbag deployment mechanisms, hypersonic vehicles, and active impact mitigation systems. Solutions will be developed to learn the dynamics at the data rates required to enable real-time decision-making systems such as those used for active control and adaptive operations. These solutions are designed for direct integration into sub-second systems to increase their resilience, robustness, safety, and viability. It follows that this research will impact society by enabling sub-second systems and empowering decision-making capabilities at speeds never reached before. Several undergraduate students will be included in the project with an emphasis on providing research experiences to underrepresented, first-generation, and low-income students by leveraging existing and valuable resources at both the University of South Carolina and Iowa State University. This project will also produce two multidisciplinary Ph.D. students with expertise in machine learning, high-rate dynamics, and control. The novelty of the approach taken in this project is to tune hyper-parameters to facilitate the use of an array of concurrent models to hide training latency. More specifically, field programmable gate arrays (FPGAs) are used to store and update the parameters of multiple concurrent long short-term memory networks as well as embed physical knowledge at the neurons' input level. This will require that the machine learning algorithm learn the temporal dependencies across operating regimes and adapt to varying dynamics. The resulting algorithm is a novel type of long short-term memory recurrent neural network that enables the prediction of nonlinear and non-stationary time series. Multiple iterations of this algorithm will be run in parallel on a single FPGA where the training time of one algorithm can be effectively hidden by another algorithm performing inference in parallel. The formulated algorithm will advance the field of real-time machine learning by furthering knowledge on: 1) how parallel models interact to hide training latency; 2) the effect of automated tuning of model parameters; 3) the role of physical knowledge in designing input spaces; 4) the benefits of subdividing non-stationary time series into local stationary systems; and 5) sustaining sufficient accuracy while meeting real-time constraints in the micro-second realm.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

该项目开发和评估用于实现实时机器学习的新框架;也就是说，对于产生大量数据的给定目标应用程序，如何处理这些数据以同时预测接下来会发生什么，同时以与目标应用程序相同的速度从过去的数据中学习。所开发的框架将产生适合于预测亚秒系统中复杂动态行为的自适应模型。 这些系统包括自适应气囊展开机构、高超音速飞行器和主动碰撞缓解系统。将开发解决方案，以所需的数据速率学习动态，以实现实时决策系统，例如用于主动控制和自适应操作的系统。 这些解决方案旨在直接集成到亚秒级系统中，以提高其弹性，鲁棒性，安全性和可行性。因此，这项研究将通过实现亚秒级系统并以前所未有的速度赋予决策能力来影响社会。几个本科生将被列入该项目，重点是通过利用南卡罗来纳州和爱荷华州州立大学的现有和宝贵的资源，为代表性不足的第一代和低收入学生提供研究经验。 该项目还将产生两个多学科博士学位。在机器学习，高速动力学和控制方面具有专业知识的学生。该项目中采用的方法的新奇在于调整超参数，以便于使用一系列并发模型来隐藏训练延迟。 更具体地说，现场可编程门阵列（FPGA）用于存储和更新多个并发长短期记忆网络的参数，以及在神经元的输入级嵌入物理知识。这将需要机器学习算法学习跨操作机制的时间依赖性，并适应变化的动态。由此产生的算法是一种新型的长短期记忆递归神经网络，使非线性和非平稳时间序列的预测。该算法的多次迭代将在单个FPGA上并行运行，其中一个算法的训练时间可以被并行执行推理的另一个算法有效地隐藏。公式化的算法将通过进一步了解以下知识来推进实时机器学习领域：1）并行模型如何相互作用以隐藏训练延迟; 2）模型参数自动调整的效果; 3）物理知识在设计输入空间中的作用; 4）将非平稳时间序列细分为局部平稳系统的好处;该奖项反映了NSF的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

项目成果

Multi-step ahead state estimation with hybrid algorithm for high-rate dynamic systems

• DOI: 10.1016/j.ymssp.2022.109536
• 发表时间: 2023-01
• 期刊: Mechanical Systems and Signal Processing
• 影响因子: 8.4
• 作者: Matthew Nelson;Vahid Barzegar;S. Laflamme;Chao Hu;A. R. Downey;Jason D. Bakos;Adam Thelen;Jacob Dodson

Deterministic and low-latency time-series forecasting of nonstationary signals

非平稳信号的确定性和低延迟时间序列预测

• DOI: 10.1117/12.2629025
• 发表时间: 2022
• 期刊: Apr. 2022
• 作者: Chowdhury, Puja;Barzegar, Vahid;Satme, Joud;Downey, Austin;Laflamme, Simon;Bakos, Jason D.;Hu, Chao
• 通讯作者: Hu, Chao

Ensemble of recurrent neural networks with long short-term memory cells for high-rate structural health monitoring

• DOI: 10.1016/j.ymssp.2021.108201
• 发表时间: 2021
• 期刊: Mechanical Systems and Signal Processing
• 影响因子: 8.4
• 作者: Vahid Barzegar;S. Laflamme;Chao Hu;J. Dodson

Sliding Mode Observer with Ensemble Learning for State Estimation of High-Rate Dynamic Systems

用于高速动态系统状态估计的集成学习滑模观测器

• 发表时间: 2021
• 期刊: IWSHM
• 作者: Nelson, M.;Wiersema, C.;Barzegar, V.;Laflamme, S.;Hu, C.;Downey, A.;Bakos, J.;Dodson, J.
• 通讯作者: Dodson, J.

Ensemble of Multi-time Resolution Recurrent Neural Networks for Enhanced Feature Extraction in High-Rate Time Series

用于增强高速时间序列特征提取的多时间分辨率循环神经网络集合

• DOI: 10.1007/978-3-030-77135-5_24
• 发表时间: 2021
• 期刊: IMAC
• 作者: Barzegar, V.;Laflamme, S.;Hu, C.;Dodson, J.
• 通讯作者: Dodson, J.