In this talk I would like to discuss some of my new and ongoing work in computational finance, machine learning and multi-language design in C++ and Python. The talk is broken down into three related topics:
1. Defining the problem, for example analytic and PDE models (e.g. (rough) Heston.)
2. Pricing models based on modern/new finite difference methods and artificial neural networks (ANNs).
3. Creating flexible software frameworks in C++ and Python that implement the algorithms from activity 2.
The advantage of this approach is that we can customise models, design frameworks and even programming languages to suit different kinds of requirements and applications.
Machine learning has changed the world of technology in the last few years by enabling the construction of predictive models from historical data. Can machine learning help with COVID-19? In this talk, I'll provide a brief overview of current therapeutic development efforts towards COVID-19 and review some of the known biology and chemistry of SARS-CoV-2. I'll then introduce the field of molecular machine learning and open source tools developed by the DeepChem project for molecular design. I'll conclude by discussing how deep learning could help design COVID-19 therapeutics.
In this talk we investigate how Deep Hedging brings a new impetus into the modelling of financial markets. While a DNN-based data-driven market generation unveils a new and highly flexible way of modelling financial time series, it is by no means "model-free". In fact, the concrete modelling choice is decisive for the features of the resulting generative model. After a very short walk through historical market models we proceed to neural network based generative modelling approaches for financial time series. We then investigate some of the challenges to achieve good results in the latter, and highlight some applications and pitfalls.
In this webinar we introduce General Stochastic Volatility models, especially new SABR/ZABR type models. These models can be tackled using Finite Difference methods or approximation formulas. Another approach is to use neural networks. Pricing and calibration is considered. Especially we introduce the CV (control variates) method applied to neural networks.
Cognitive/AI systems process knowledge that is far too complex for current databases. They require an expressive data model and an intelligent query language to perform knowledge engineering over complex datasets.
In this talk, we will discuss how Grakn, a database to organise complex networks of data and make it queryable, provides the knowledge graph foundation for intelligent systems to manage complex data.
Four quantitative analyses and simulations to find the Bitcoin fair value price and forecast its potential growth in the coming years.
We will test Stock to Flow model, Rate of Adoption model, Hash Rate Remuneration model and Monte Carlo Simulations.
Using next-generation artificial intelligence and deep-learning techniques, Insilico Medicine identifies novel molecular structures with unique properties for drug discovery programmes in different disease areas, as well as for crucial Longevity research.