Generative Models for Stochastic Processes Using Convolutional Neural Networks

TL;DR

This paper explores using convolutional neural networks as a flexible generative approach for stochastic processes, allowing diverse fields to perform forecasts and simulations without detailed system assumptions.

Contribution

It introduces a novel application of CNNs as a general tool for modeling stochastic processes, bypassing the need for explicit system identification or parameter estimation.

Findings

CNNs can effectively generate stochastic process data

The approach is versatile across different scientific fields

It simplifies modeling without detailed system assumptions

Abstract

The present paper aims to demonstrate the usage of Convolutional Neural Networks as a generative model for stochastic processes, enabling researchers from a wide range of fields (such as quantitative finance and physics) to develop a general tool for forecasts and simulations without the need to identify/assume a specific system structure or estimate its parameters.