Hong Kong Machine Learning Season 4 Episode 7

 13.04.2022 -  Hong Kong Machine Learning -  ~3 Minutes


  • Wednesday, April 13, 2022 from 7:00 PM to 9:00 PM (Hong Kong Time)


  • This meetup was hosted on zoom.

The page of the event on Meetup: HKML S4E7


Talk 1: Recent Advances in Machine Learning Applications, and Trends

Speaker: Gautier Marti (10min)

More of an introduction to the Hong Kong Machine Learning Meetups, and a call for future talks than a focused presentation. Not technical (beginner level). The goal is to highlight applications of ML in environmental sciences, journalism, healthcare, finance, and history (social sciences). We hope to have more technical talks (intermediate to advanced level) at future Meetups on these Machine Learning applications by those who are advancing the knowledge frontiers. Please, reach out if you work in these areas, or if you know people who are involved in similar applications.


Talk 2: Pricing options on flow forwards by neural networks in Hilbert space

Speaker: Luca Galimberti (40min)

Abstract: We propose a new methodology for pricing options on flow forwards by applying infinite-dimensional neural networks. We recast the pricing problem as an optimization problem in a Hilbert space of real-valued function on the positive real line, which is the state space for the term structure dynamics. This optimization problem is solved by facilitating a novel feedforward neural network architecture designed for approximating continuous functions on the state space. The proposed neural net is built upon the basis of the Hilbert space. We provide an extensive case study that shows excellent numerical efficiency, with superior performance over that of a classical neural net trained on sampling the term structure curves.

paper: https://arxiv.org/pdf/2202.11606.pdf


Lightning Talk 3: Designing a crypto bot infrastructure

Speaker: Chris Kang (15min)

Unlike the traditional stock markets, the cryptocurrency market never sleeps. Therefore, it is crucial to have tools to manage investment accessible. Therefore, I have designed a course at HKML EduTech (https://www.hkml-edutech.com) on how to design and program an infrastructure for a crypto bot from scratch using Python and Telegram. Chapter 1 is about understanding crypto exchange API documentation and utilizing the endpoints provided by the exchange. Chapter 2 is about setting up a telegram bot and designing the bot’s function. Chapter 3 is about setting up AWS EC2 so that the bot runs without disconnection. Lastly, chapter 4 is about integrating the previous 3 chapters. In addition, my personal experience/project with the bots and the other potential of the bot will be presented (with hints to some machine learning applications).

Register your interest for this course there, or contact us directly here.

Video Recording of the HKML Meetup on YouTube

  • YouTube videos:

HKML S4E7 - Pricing options on flow forwards by neural networks in Hilbert space

HKML S4E7 - Pricing options on flow forwards by neural networks in Hilbert space