New Machine Learning Architectures For Fast Data Selection

用于快速数据选择的新机器学习架构

• 批准号: ST/X005089/1
• 负责人: Sudarshan Paramesvaran
• 金额: $ 3.14万
• 依托单位: University of Bristol
• 依托单位国家: 英国
• 项目类别: Research Grant
• 财政年份: 2022
• 资助国家: 英国
• 起止时间: 2022 至 无数据
• 项目状态: 已结题
• 来源: https://gtr.ukri.org/projects?ref=ST%2FX005089%2F1
• 关键词: New; Machine; Learning; Architectures; Fast

Particle Physics experiments can have huge limitations to their Physics reach due to the sheer amount of data that is produced, and consequently how little they can store. In addition huge amounts of the data produced at a facility such as the LHC are intrinsically not as scientifically interesting as rare processes. In order to solve these challenges "Trigger" systems were developed to quickly scan data and identify potentially interesting signatures. These systems have to maintain a high efficiency for the processes of interest while also staying in the technical confines of the experiment such as latency - the time it takes for the algorithm to make its decision. Several technologies have been used in this endeavour such as FPGAs, CPUs, or GPUs at various stages of selection, with each of the solutions better suited to different tasks. However each of the solutions have their own limitations. The new Xilinx Versal platform aims to combine the benefits of CPUs, GPUs and FPGAs into a flexible platform for parallel processing. This architecture is particularly well suited to machine learning (ML) based algorithms and we propose to exploit our expertise in this area to evaluate these new platforms and gain a head-start in developing new ML architectures for future particle physics experiments

粒子物理学实验可能会对物理学范围产生巨大的限制，因为产生的数据量很大，因此它们可以存储的数据很少。此外，在大型强子对撞机这样的设施中产生的大量数据本质上并不像罕见的过程那样具有科学意义。为了解决这些挑战，开发了“触发”系统来快速扫描数据并识别潜在的感兴趣的签名。这些系统必须保持感兴趣的过程的高效率，同时也保持在实验的技术限制，如延迟-算法做出决定所需的时间。在这方面已经使用了几种技术，例如FPGA，CPU或GPU在不同的选择阶段，每种解决方案都更适合不同的任务。然而，每种解决方案都有其自身的局限性。新的Xilinx Versal平台旨在将CPU、GPU和FPGA的优势联合收割机结合到一个灵活的并行处理平台中。这种架构特别适合基于机器学习（ML）的算法，我们建议利用我们在这一领域的专业知识来评估这些新平台，并在为未来的粒子物理实验开发新的ML架构方面取得领先优势