The signature approach for the supervised learning problem on the path space and its application

Inspired by the rough paths theory (RPT) the core object of RPT – the so-called the signature of the path provides a good feature set for sequential data.  In the talk, we discuss how to combine the recurrent neural network with the signature feature set to tackle the supervised learning problem on the path space, which is based on the theory of the solution to the stochastic differential equation (SDE). We will apply this method to learn the solution to unknown SDEs without any prior knowledge and demonstrate the effectiveness of this method. Finally we will discuss the potential applications of this method.

Introduction

Dr. Ni a senior lecturer in financial mathematics at University College London (UCL) and the Turing Fellow at the Alan Turing Institute since September 2016. Prior to this she was a visiting postdoctoral researcher at ICERM and Department of Applied Mathematics at Brown University from 2012/09 to 2013/05 and continued her postdoctoral research at the Oxford-Man Institute of Quantitative Finance until 2016.  Dr. Ni finished her D.Phil. in mathematics in 2012 under the supervision of Professor Terry Lyons at University of Oxford.

Her research interests include stochastic analysis, financial mathematics and machine learning. More specifically she is interested in non-parametric modelling effects of data streams through rough paths theory and statistical models. Rough paths theory is a non-linear extension of classical theory of control differential equations to model highly oscillatory systems, and the core concept in rough paths theory is the signature of a path, which can be used as useful features for learning to summarize sequential data in terms of its effect. Moreover, she is also interested in its applications, e.g. online Chinese handwritten character and financial data streams.