南昌大学百年校庆系列学术报告 理学院格物论坛（二十二）王瑜研究员学术报告

题目:  Machine learning for astrophysics: inferring redshift, testing physical model and others

报告人： 王瑜 研究员 浙江工业大学

报告时间： 2021年4月15日（周四）下午 14:00

报告地点： 理生楼B310

报告人简介：

王瑜现为浙江工业大学客座研究员，他也是意大利国际相对论天体物理学网络中心（ICRANet）的研究员，在罗马大学和意大利国家天体物理研究所（INAF）担任研究职务。主要从事高能天体物理理论和数据研究，特别是中子星、黑洞和伽马射线暴。王瑜目前专注于天体物理学的机器学习，通过神经网络来研究活动星系核、宇宙学、伽马射线暴和引力波等。不同于传统的通过提出理论或总结观测来理解自然，引入深度学习以机器视角看大数据，将物理规律转化为其直觉，人类可以通过理解机器来理解自然。

报告内容简介:

The advancements of observational detectors and supercomputers have led to the immense growth of astronomical and simulated data, which brings the feasibility to train machines to seek and to recognize the common characteristics, and in turn accelerates the simulation, assists the observation and infers the physics. In this talk, I will first introduce a basic application that optimizes a neural network to tell the redshift of a quasar from its SDSS optical spectrum. Then I will show a further example of a machine working for physics, which test the unification model of active galactic nucleus (AGN) by a neural network. Machine learning has been widely used for a variety of tasks. I will briefly exhibit some applications on cosmology, gravitational wave and gamma-ray burst for the rest of the talk.

欢迎广大师生参加！

南昌大学科学技处

南昌大学理学院物理系