Explore the intersection of neural stochastic differential equations, deep learning, and stochastic control in this lecture by Lukasz Szpruch from the Alan Turing Institute and University of Edinburgh. Delve into modeling techniques in finance, comparing classical models with modern approaches like optimal transport and neural networks. Learn about calibration algorithms, simulations, and extensions incorporating additional market information. Examine the connections between neural SDEs, stochastic control, and gradient flows, with practical examples including stochastic gradient descent. Gain insights into cutting-edge quantitative finance techniques as part of the Fields-CFI Bootcamp on Machine Learning for Quantitative Finance.
Neural SDEs, Deep Learning and Stochastic Control
Overview
Syllabus
Introduction
Deep Learning and Stochastic Control
Modeling in Finance
Classical Models
Disadvantages
Optimal Transport
Neural Networks
Calibration
Algorithm
Simulations
Extensions
Additional Market Information
Neural SDEs
Stochastic Control
Gradient Flows
Example
Stochastic Gradient Descent
Taught by
Fields Institute
