Deep learning: the first billion years with next generation Telescopes

深度学习：下一代望远镜的第一个十亿年

• 批准号: FL180100060
• 负责人: Prof Karl Glazebrook
• 金额: $ 200.91万
• 依托单位: Swinburne University of Technology
• 依托单位国家: 澳大利亚
• 项目类别: Australian Laureate Fellowships
• 财政年份: 2019
• 资助国家: 澳大利亚
• 起止时间: 2019-08-05 至 2025-08-03
• 项目状态: 未结题
• 来源: https://dataportal.arc.gov.au/RGS/Web/Grants/FL180100060
• 关键词: Deep; learning; first; billion; years

This project aims to transform our understanding of the early Universe using the giant James Webb Space Telescope. This telescope, which will be the most transformative telescope of the next decade will be launched in early 2019. The telescope will observe the dawn of galaxy formation 13 billion years ago, a time that is currently shrouded in obscurity. The project will develop new techniques in scientific computing, including cloud-based workflows to make science more efficient and the application of 'deep learning' to the discovery of new astronomical objects and the acceleration of scientific computation. The project will train a new generation of young scientists in the use of these techniques, and actively engage with school students.

该项目旨在利用巨大的詹姆斯韦伯太空望远镜改变我们对早期宇宙的理解。这台望远镜将成为未来十年最具变革性的望远镜，将于2019年初发射。该望远镜将观测130亿年前星系形成的曙光，这段时间目前还笼罩在朦胧之中。该项目将开发科学计算方面的新技术，包括基于云的工作流程，以提高科学效率，并将“深度学习”应用于发现新的天文物体和加速科学计算。该项目将培训新一代青年科学家使用这些技术，并积极与学生接触。