[Online] Hong Kong Machine Learning Meetup Season 4 Episode 6

Back to online Meetups on Zoom for the foreseeable future in Hong Kong. No one is willing to spend some time in Penny's Bay if it can be avoided.

Talk 1: Sequential Bootstrapping in Finance: Approaching the true IID Sampling

Speaker: Valeriia Pervushyna, Quantitative Researcher at Hudson & Thames https://hudsonthames.org

Abstract: Most classic Machine Learning approaches highly rely on the assumption of the i.i.d of samples. However, in the case of financial applications, achieving the first "I" - independence is close to impossible if using the standard ML approaches. Sequential bootstrapping allows to maximize the "purity" of the obtained samples in terms of independence and achieve results closest to the true IID sampling.

Talk 2: From galaxy pairs to galaxy mergers

Speaker: Hugo Pfister, Postdoctoral Researcher, The University of Hong Kong

Machine Learning entered in almost every aspect of our life, and research in Astrophysics is not an exception! In this talk, I will begin with a brief introduction of the astrophysical problems of galaxy mergers: what are they? why are they interesting? what is our current understanding? In the second part, I will present how I used a combination of “cosmological simulations” and machine learning to automate the detection of galaxy mergers.

lightning talk: TBA