An unsupervised deep learning approach to solving partial integro-differential equations

@article{Hirsa2022AnUD,
  title={An unsupervised deep learning approach to solving partial integro-differential equations},
  author={Ali Hirsa and Weilong Fu},
  journal={Quantitative Finance},
  year={2022},
  volume={22},
  pages={1481 - 1494}
}
We investigate solving partial integro-differential equations (PIDEs) using unsupervised deep learning. The PIDE is employed for option pricing, when the underlying process is a Lévy process with jumps. The unsupervised deep learning approach employs a neural network as the candidate solution and trains the neural network to satisfy the PIDE. By matching the PIDE and the boundary conditions, the neural network would yield an accurate solution to the PIDE. Unlike supervised learning, this… 

Neural Integro-Differential Equations

TLDR
The Neural IDE (NIDE), a novel deep learning framework based on the theory of IDEs where integral operators are learned using neural networks is introduced and it is demonstrated that NIDE outperforms other models.

An approximate solution for options market-making in high dimension

TLDR
This paper shows that a simple analytical approximation of the solution of the market maker's problem provides significantly higher flexibility than the existing algorithms designing options market making strategies.

Solving barrier options under stochastic volatility using deep learning

We develop an unsupervised deep learning method to solve the barrier options under the Bergomi model. The neural networks serve as the approximate option surfaces and are trained to satisfy the PDE

References

SHOWING 1-10 OF 46 REFERENCES

DGM: A deep learning algorithm for solving partial differential equations

Solving high-dimensional partial differential equations using deep learning

TLDR
A deep learning-based approach that can handle general high-dimensional parabolic PDEs using backward stochastic differential equations and the gradient of the unknown solution is approximated by neural networks, very much in the spirit of deep reinforcement learning with the gradient acting as the policy function.

Artificial neural networks for solving ordinary and partial differential equations

TLDR
This article illustrates the method by solving a variety of model problems and presents comparisons with solutions obtained using the Galekrkin finite element method for several cases of partial differential equations.

Deep learning calibration of option pricing models: some pitfalls and solutions

TLDR
This paper considers a classical problem of mathematical finance - calibration of option pricing models to market data, as it was recently drawn some attention of the financial society in the context of deep learning and artificial neural networks.

Neural-network methods for boundary value problems with irregular boundaries

TLDR
Two networks are employed: a multilayer perceptron and a radial basis function network to account for the exact satisfaction of the boundary conditions of complex boundary geometry.

Supervised Deep Neural Networks (DNNs) for Pricing/Calibration of Vanilla/Exotic Options Under Various Different Processes

TLDR
It is demonstrated that deep neural networks exponentially expedite option pricing compared to commonly used option pricing methods which consequently make calibration and parameter estimation super fast.

A neural network-based framework for financial model calibration

TLDR
The rapid on-line learning of implied volatility by ANNs, in combination with the use of an adapted parallel global optimization method, tackles the computation bottleneck and provides a fast and reliable technique for calibrating model parameters while avoiding, as much as possible, getting stuck in local minima.

Deep learning volatility: a deep neural network perspective on pricing and calibration in (rough) volatility models

TLDR
A neural network-based calibration method that performs the calibration task within a few milliseconds for the full implied volatility surface, and stochastic volatility models (classical and rough) can be handled in great generality as the framework also allows taking the forward variance curve as an input.

A Fast and Accurate FFT-Based Method for Pricing Early-Exercise Options under L[e-acute]vy Processes

TLDR
The main idea is to reformulate the well-known risk-neutral valuation formula by recognizing that it is a convolution, and the resulting convolution is dealt with numerically by using the fast Fourier transform.

Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift

TLDR
Applied to a state-of-the-art image classification model, Batch Normalization achieves the same accuracy with 14 times fewer training steps, and beats the original model by a significant margin.