Solving high-dimensional partial differential equations using deep learning
Abstract
Developing algorithms for solving high-dimensional partial differential equations (PDEs) has been an exceedingly difficult task for a long time, due to the notoriously difficult problem known as the “curse of dimensionality.” This paper introduces a deep learning-based approach that can handle general high-dimensional parabolic PDEs. To this end, the PDEs are reformulated 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. Numerical results on examples including the nonlinear Black–Scholes equation, the Hamilton–Jacobi–Bellman equation, and the Allen–Cahn equation suggest that the proposed algorithm is quite effective in high dimensions, in terms of both accuracy and cost. Furthermore, this opens up possibilities in economics, finance, operational research, and physics, by considering all participating agents, assets, resources, or particles together at the same time, instead of making ad hoc assumptions on their interrelationships.
- Authors:
-
[1];
- Princeton Univ., Princeton, NJ (United States)
- ETH Zurich, Zurich (Switzerland)
- Princeton Univ., Princeton, NJ (United States); Beijing Inst. of Big Data Research, Beijing (China)
- Publication Date:
- Research Org.:
- Princeton Univ., NJ (United States)
- Sponsoring Org.:
- USDOE Office of Science (SC)
- OSTI Identifier:
- 1540276
- Grant/Contract Number:
- SC0009248
- Resource Type:
- Accepted Manuscript
- Journal Name:
- Proceedings of the National Academy of Sciences of the United States of America
- Additional Journal Information:
- Journal Volume: 115; Journal Issue: 34; Journal ID: ISSN 0027-8424
- Publisher:
- National Academy of Sciences
- Country of Publication:
- United States
- Language:
- English
- Subject:
- 97 MATHEMATICS AND COMPUTING; Science & Technology; Other Topics; partial differential equations; backward stochastic differential equations; high dimension; deep learning; Feynman–Kac
Citation Formats
Han, Jiequn, Jentzen, Arnulf, and E, Weinan. Solving high-dimensional partial differential equations using deep learning. United States: N. p., 2018.
Web. doi:10.1073/pnas.1718942115.
Han, Jiequn, Jentzen, Arnulf, & E, Weinan. Solving high-dimensional partial differential equations using deep learning. United States. https://doi.org/10.1073/pnas.1718942115
Han, Jiequn, Jentzen, Arnulf, and E, Weinan. Mon .
"Solving high-dimensional partial differential equations using deep learning". United States. https://doi.org/10.1073/pnas.1718942115. https://www.osti.gov/servlets/purl/1540276.
@article{osti_1540276,
title = {Solving high-dimensional partial differential equations using deep learning},
author = {Han, Jiequn and Jentzen, Arnulf and E, Weinan},
abstractNote = {Developing algorithms for solving high-dimensional partial differential equations (PDEs) has been an exceedingly difficult task for a long time, due to the notoriously difficult problem known as the “curse of dimensionality.” This paper introduces a deep learning-based approach that can handle general high-dimensional parabolic PDEs. To this end, the PDEs are reformulated 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. Numerical results on examples including the nonlinear Black–Scholes equation, the Hamilton–Jacobi–Bellman equation, and the Allen–Cahn equation suggest that the proposed algorithm is quite effective in high dimensions, in terms of both accuracy and cost. Furthermore, this opens up possibilities in economics, finance, operational research, and physics, by considering all participating agents, assets, resources, or particles together at the same time, instead of making ad hoc assumptions on their interrelationships.},
doi = {10.1073/pnas.1718942115},
journal = {Proceedings of the National Academy of Sciences of the United States of America},
number = 34,
volume = 115,
place = {United States},
year = {Mon Aug 06 00:00:00 EDT 2018},
month = {Mon Aug 06 00:00:00 EDT 2018}
}
Free Publicly Available Full Text
Publisher's Version of Record
Other availability
Search WorldCat to find libraries that may hold this journal
Cited by: 638 works
Citation information provided by
Web of Science
Web of Science
Save to My Library
You must Sign In or Create an Account in order to save documents to your library.
Works referenced in this record:
Option Pricing and Replication with Transactions Costs
journal, December 1985
- Leland, Hayne E.
- The Journal of Finance, Vol. 40, Issue 5
A numerical algorithm for a class of BSDEs via the branching process
journal, February 2014
- Henry-Labordère, Pierre; Tan, Xiaolu; Touzi, Nizar
- Stochastic Processes and their Applications, Vol. 124, Issue 2
Approximate Dynamic Programming: Solving the Curses of Dimensionality
book, August 2011
- Powell, Warren B.
- Wiley Series in Probability and Statistics
Forward-backward stochastic differential equations and quasilinear parabolic PDEs
journal, May 1999
- Pardoux, Etienne; Tang, Shanjian
- Probability Theory and Related Fields, Vol. 114, Issue 2
Deep learning
journal, May 2015
- LeCun, Yann; Bengio, Yoshua; Hinton, Geoffrey
- Nature, Vol. 521, Issue 7553
Algorithms for overcoming the curse of dimensionality for certain Hamilton–Jacobi equations arising in control theory and elsewhere
journal, September 2016
- Darbon, Jérôme; Osher, Stanley
- Research in the Mathematical Sciences, Vol. 3, Issue 1
A Primal-Dual Algorithm for Bsdes: a Primal-Dual Algorithm for Bsdes
journal, June 2015
- Bender, Christian; Schweizer, Nikolaus; Zhuo, Jia
- Mathematical Finance, Vol. 27, Issue 3
The Pricing of Options and Corporate Liabilities
journal, May 1973
- Black, Fischer; Scholes, Myron
- Journal of Political Economy, Vol. 81, Issue 3
Recursive valuation of defaultable securities and the timing of resolution of uncertainty
journal, November 1996
- Duffie, Darrell; Schroder, Mark; Skiadas, Costis
- The Annals of Applied Probability, Vol. 6, Issue 4
Mastering the game of Go with deep neural networks and tree search
journal, January 2016
- Silver, David; Huang, Aja; Maddison, Chris J.
- Nature, Vol. 529, Issue 7587
Adaptive importance sampling in least-squares Monte Carlo algorithms for backward stochastic differential equations
journal, April 2017
- Gobet, E.; Turkedjiev, P.
- Stochastic Processes and their Applications, Vol. 127, Issue 4
Approximation theory of the MLP model in neural networks
journal, January 1999
- Pinkus, Allan
- Acta Numerica, Vol. 8
Pricing and hedging derivative securities in markets with uncertain volatilities
journal, June 1995
- Avellaneda, M.; Levy ∗, A.; ParÁS, A.
- Applied Mathematical Finance, Vol. 2, Issue 2
The randomized information complexity of elliptic PDE
journal, April 2006
- Heinrich, Stefan
- Journal of Complexity, Vol. 22, Issue 2
Option Pricing with Differential Interest Rates
journal, April 1995
- Bergman, Yaacov Z.
- Review of Financial Studies, Vol. 8, Issue 2
Deep Neural Networks for Acoustic Modeling in Speech Recognition: The Shared Views of Four Research Groups
journal, November 2012
- Hinton, Geoffrey; Deng, Li; Yu, Dong
- IEEE Signal Processing Magazine, Vol. 29, Issue 6
Counterparty risk and Funding: the four Wings of the tva
journal, March 2013
- CrÉPey, StÉPhane; Gerboud, RÉMi; Grbac, Zorana
- International Journal of Theoretical and Applied Finance, Vol. 16, Issue 02
Backward Stochastic Differential Equations in Finance
journal, January 1997
- El Karoui, N.; Peng, S.; Quenez, M. C.
- Mathematical Finance, Vol. 7, Issue 1
Counterparty Risk Valuation: A Marked Branching Diffusion Approach
journal, January 2012
- Henry-Labordere, Pierre
- SSRN Electronic Journal
Implicit solution of uncertain volatility/transaction cost option pricing models with discretely observed barriers
journal, March 2001
- Forsyth, P. A.; Vetzal, K. R.
- Applied Numerical Mathematics, Vol. 36, Issue 4
Approximate Dynamic Programming: Solving the Curses of Dimensionality
book, January 2007
- Powell, Warren B.
Counterparty Risk Valuation: A Marked Branching Diffusion Approach
preprint, January 2012
- Henry-Labordere, Pierre
- arXiv
Counterparty Risk and Funding: The Four Wings of the TVA
preprint, January 2012
- Crépey, Stéphane; Gerboud, Rémi; Grbac, Zorana
- arXiv
Algorithms for Overcoming the Curse of Dimensionality for Certain Hamilton-Jacobi Equations Arising in Control Theory and Elsewhere
preprint, January 2016
- Darbon, Jérôme; Osher, Stanley
- arXiv
Option Pricing with Differential Interest Rates
journal, April 1995
- Bergman, Yaacov Z.
- Review of Financial Studies, Vol. 8, Issue 2
Option Pricing and Replication with Transactions Costs
journal, December 1985
- Leland, Hayne E.
- The Journal of Finance, Vol. 40, Issue 5
Works referencing / citing this record:
A Deeper Look into ‘Deep Learning of Aftershock Patterns Following Large Earthquakes’: Illustrating First Principles in Neural Network Physical Interpretability
book, May 2019
- Mignan, Arnaud; Broccardo, Marco; Rojas, Ignacio
- Advances in Computational Intelligence: 15th International Work-Conference on Artificial Neural Networks, IWANN 2019, Gran Canaria, Spain, June 12-14, 2019, Proceedings, Part I, p. 3-14
Neural networks catching up with finite differences in solving partial differential equations in higher dimensions
journal, January 2020
- Avrutskiy, Vsevolod I.
- Neural Computing and Applications, Vol. 32, Issue 17
Multiscale Modeling Meets Machine Learning: What Can We Learn?
journal, February 2020
- Peng, Grace C. Y.; Alber, Mark; Buganza Tepole, Adrian
- Archives of Computational Methods in Engineering
Solving partial differential equations by a supervised learning technique, applied for the reaction–diffusion equation
journal, November 2019
- Zakeri, Behzad; Khashehchi, Morteza; Samsam, Sanaz
- SN Applied Sciences, Vol. 1, Issue 12
Data-driven acceleration of photonic simulations
journal, December 2019
- Trivedi, Rahul; Su, Logan; Lu, Jesse
- Scientific Reports, Vol. 9, Issue 1
Integrating machine learning and multiscale modeling—perspectives, challenges, and opportunities in the biological, biomedical, and behavioral sciences
journal, November 2019
- Alber, Mark; Buganza Tepole, Adrian; Cannon, William R.
- npj Digital Medicine, Vol. 2, Issue 1
Solving Fokker-Planck equation using deep learning
journal, January 2020
- Xu, Yong; Zhang, Hao; Li, Yongge
- Chaos: An Interdisciplinary Journal of Nonlinear Science, Vol. 30, Issue 1
Using machine learning to predict extreme events in complex systems
journal, December 2019
- Qi, Di; Majda, Andrew J.
- Proceedings of the National Academy of Sciences, Vol. 117, Issue 1
Machine-learning solver for modified diffusion equations
journal, November 2018
- Wei, Qianshi; Jiang, Ying; Chen, Jeff Z. Y.
- Physical Review E, Vol. 98, Issue 5
Numerical Solution of a Class of Nonlinear Partial Differential Equations by Using Barycentric Interpolation Collocation Method
journal, December 2018
- Wu, Hongchun; Wang, Yulan; Zhang, Wei
- Mathematical Problems in Engineering, Vol. 2018
Variational Monte Carlo -- Bridging concepts of machine learning and high dimensional partial differential equations
text, January 2018
- Eigel, Martin; Trunschke, Philipp; Schneider, Reinhold
- Weierstrass Institute
Variational Monte Carlo—bridging concepts of machine learning and high-dimensional partial differential equations
journal, October 2019
- Eigel, Martin; Schneider, Reinhold; Trunschke, Philipp
- Advances in Computational Mathematics, Vol. 45, Issue 5-6
A machine-learning solver for modified diffusion equations
text, January 2018
- Wei, Qianshi; Jiang, Ying; Chen, Jeff Z. Y.
- arXiv
Variational Monte Carlo - Bridging Concepts of Machine Learning and High Dimensional Partial Differential Equations
text, January 2018
- Eigel, Martin; Schneider, Reinhold; Trunschke, Philipp
- arXiv
Data-driven acceleration of Photonic Simulations
preprint, January 2019
- Trivedi, Rahul; Su, Logan; Lu, Jesse
- arXiv
Solving Fokker-Planck equation using deep learning
text, January 2019
- Xu, Yong; Zhang, Hao; Li, Yongge
- arXiv
Multiscale modeling meets machine learning: What can we learn?
preprint, January 2019
- Peng, Grace C. Y.; Alber, Mark; Tepole, Adrian Buganza
- arXiv
A Theoretical Analysis of Deep Neural Networks and Parametric PDEs
journal, June 2021
- Kutyniok, Gitta; Petersen, Philipp; Raslan, Mones
- Constructive Approximation, Vol. 55, Issue 1
SelectNet: Self-paced learning for high-dimensional partial differential equations
journal, September 2021
- Gu, Yiqi; Yang, Haizhao; Zhou, Chao
- Journal of Computational Physics, Vol. 441
Using machine learning to predict extreme events in complex systems
journal, December 2019
- Qi, Di; Majda, Andrew J.
- Proceedings of the National Academy of Sciences, Vol. 117, Issue 1
Overcoming the curse of dimensionality in the numerical approximation of semilinear parabolic partial differential equations
journal, December 2020
- Hutzenthaler, Martin; Jentzen, Arnulf; Kruse, Thomas
- Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences, Vol. 476, Issue 2244
Neural network solutions to differential equations in nonconvex domains: Solving the electric field in the slit-well microfluidic device
journal, July 2020
- Magill, Martin; Nagel, Andrew M.; de Haan, Hendrick W.
- Physical Review Research, Vol. 2, Issue 3
Web-enabled Intelligent System for Continuous Sensor Data Processing and Visualization
conference, July 2019
- Hamza-Lup, Felix G.; Iacob, Ionut E.; Khan, Sushmita
- The 24th International Conference on 3D Web Technology
Pricing Options and Computing Implied Volatilities using Neural Networks
journal, February 2019
- Liu, Shuaiqiang; Oosterlee, Cornelis; Bohte, Sander
- Risks, Vol. 7, Issue 1
PowerNet: Efficient Representations of Polynomials and Smooth Functions by Deep Neural Networks with Rectified Power Units
journal, June 2020
- Li, Bo
- Journal of Mathematical Study, Vol. 53, Issue 2
A Local Deep Learning Method for Solving High Order Partial Differential Equations
journal, June 2022
- Yang, Jiang; Zhu, Quanhui
- Numerical Mathematics: Theory, Methods and Applications, Vol. 15, Issue 1
Solving parametric PDE problems with artificial neural networks
text, January 2017
- Khoo, Yuehaw; Lu, Jianfeng; Ying, Lexing
- arXiv
An unbiased Ito type stochastic representation for transport PDEs: A Toy Example
preprint, January 2018
- Reis, Goncalo dos; Smith, Greig
- arXiv
Forward-Backward Stochastic Neural Networks: Deep Learning of High-dimensional Partial Differential Equations
preprint, January 2018
- Raissi, Maziar
- arXiv
Analysis of the Generalization Error: Empirical Risk Minimization over Deep Artificial Neural Networks Overcomes the Curse of Dimensionality in the Numerical Approximation of Black-Scholes Partial Differential Equations
text, January 2018
- Berner, Julius; Grohs, Philipp; Jentzen, Arnulf
- arXiv
Unbiased deep solvers for linear parametric PDEs
text, January 2018
- Vidales, Marc Sabate; Siska, David; Szpruch, Lukasz
- arXiv
Learning to Optimize Multigrid PDE Solvers
preprint, January 2019
- Greenfeld, Daniel; Galun, Meirav; Kimmel, Ron
- arXiv
Enforcing constraints for time series prediction in supervised, unsupervised and reinforcement learning
preprint, January 2019
- Stinis, Panos
- arXiv
Learning to Discretize: Solving 1D Scalar Conservation Laws via Deep Reinforcement Learning
preprint, January 2019
- Wang, Yufei; Shen, Ziju; Long, Zichao
- arXiv
Deep 2FBSDEs For Systems With Control Multiplicative Noise
preprint, January 2019
- Pereira, Marcus A.; Wang, Ziyi; Chen, Tianrong
- arXiv
Three algorithms for solving high-dimensional fully-coupled FBSDEs through deep learning
preprint, January 2019
- Ji, Shaolin; Peng, Shige; Peng, Ying
- arXiv
Accelerated Share Repurchase and other buyback programs: what neural networks can bring
preprint, January 2019
- Guéant, Olivier; Manziuk, Iuliia; Pu, Jiang
- arXiv
Numerical resolution of McKean-Vlasov FBSDEs using neural networks
preprint, January 2019
- Germain, Maximilien; Mikael, Joseph; Warin, Xavier
- arXiv
Towards Robust and Stable Deep Learning Algorithms for Forward Backward Stochastic Differential Equations
preprint, January 2019
- Güler, Batuhan; Laignelet, Alexis; Parpas, Panos
- arXiv
Algorithms of Data Development For Deep Learning and Feedback Design
text, January 2019
- Kang, Wei; Gong, Qi; Nakamura-Zimmerer, Tenavi
- arXiv
Solving Partial Differential Equations with Neural Networks
preprint, January 2019
- Pedro, Juan B.; Maroñas, Juan; Paredes, Roberto
- arXiv
Strong solutions of forward-backward stochastic differential equations with measurable coefficients
preprint, January 2020
- Luo, Peng; Menoukeu-Pamen, Olivier; Tangpi, Ludovic
- arXiv
Deep combinatorial optimisation for optimal stopping time problems : application to swing options pricing
preprint, January 2020
- Deschatre, Thomas; Mikael, Joseph
- arXiv
On Calibration Neural Networks for extracting implied information from American options
preprint, January 2020
- Liu, Shuaiqiang; Leitao, Álvaro; Borovykh, Anastasia
- arXiv
Learning To Solve Differential Equations Across Initial Conditions
preprint, January 2020
- Malik, Shehryar; Anwar, Usman; Ahmed, Ali
- arXiv
Backward Deep BSDE Methods and Applications to Nonlinear Problems
preprint, January 2020
- Yu, Yajie; Hientzsch, Bernhard; Ganesan, Narayan
- arXiv
Quantum Ground States from Reinforcement Learning
preprint, January 2020
- Barr, Ariel; Gispen, Willem; Lamacraft, Austen
- arXiv
Deep neural network approximations for the stable manifolds of the Hamilton-Jacobi equations
preprint, January 2020
- Chen, Guoyuan
- arXiv
Unsupervised Neural Networks for Quantum Eigenvalue Problems
preprint, January 2020
- Jin, Henry; Mattheakis, Marios; Protopapas, Pavlos
- arXiv
Solving path dependent PDEs with LSTM networks and path signatures
preprint, January 2020
- Sabate-Vidales, Marc; Šiška, David; Szpruch, Lukasz
- arXiv
Large-Scale Multi-Agent Deep FBSDEs
preprint, January 2020
- Chen, Tianrong; Wang, Ziyi; Exarchos, Ioannis
- arXiv
Novel multi-step predictor-corrector schemes for backward stochastic differential equations
preprint, January 2021
- Han, Qiang; Ji, Shaolin
- arXiv
Learning optimal multigrid smoothers via neural networks
preprint, January 2021
- Huang, Ru; Li, Ruipeng; Xi, Yuanzhe
- arXiv
Solving Backward Doubly Stochastic Differential Equations through Splitting Schemes
preprint, January 2021
- Bao, Feng; Cao, Yanzhao; Zhang, He
- arXiv
Understanding Loss Landscapes of Neural Network Models in Solving Partial Differential Equations
preprint, January 2021
- Wu, Keke; Du, Rui; Chen, Jingrun
- arXiv
On the Representation of Solutions to Elliptic PDEs in Barron Spaces
preprint, January 2021
- Chen, Ziang; Lu, Jianfeng; Lu, Yulong
- arXiv
Learning Neural Hamiltonian Dynamics: A Methodological Overview
preprint, January 2022
- Chen, Zhijie; Feng, Mingquan; Yan, Junchi
- arXiv
Sobolev Acceleration and Statistical Optimality for Learning Elliptic Equations via Gradient Descent
preprint, January 2022
- Lu, Yiping; Blanchet, Jose; Ying, Lexing
- arXiv
Probability flow solution of the Fokker-Planck equation
preprint, January 2022
- Boffi, Nicholas M.; Vanden-Eijnden, Eric
- arXiv