Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

arXiv:2512.07919

Quantum algorithms for viscosity solutions to nonlinear Hamilton-Jacobi equations based on an entropy penalisation method

Shi Jin, Nana Liu

quant-phmath-phSubmitted Dec 8, 2025• Announced Dec 9, 2025• 0 citations
v1 · Dec 8, 2025v2 · Feb 16, 2026
arXivHTMLPDF

Abstract

We present a framework for efficient extraction of the viscosity solutions of nonlinear Hamilton-Jacobi equations with convex Hamiltonians. These viscosity solutions play a central role in areas such as front propagation, mean-field games, optimal control, machine learning, and a direct application to the forced Burgers' equation. Our method is based on an entropy penalisation method proposed by Gomes and Valdinoci, which generalises the Cole-Hopf transform from quadratic to general convex Hamiltonians, allowing a reformulation of viscous Hamilton-Jacobi dynamics by a discrete-time linear dynamics which approximates a linear heat-like parabolic equation, and can also extend to continuous-time dynamics. This makes the method suitable for quantum simulation. The validity of these results hold for arbitrary nonlinearity that correspond to convex Hamiltonians, and for arbitrarily long times, thus obviating a chief obstacle in most quantum algorithms for nonlinear partial differential equations. We provide quantum algorithms, both analog and digital, for extracting pointwise values, gradients, minima, and function evaluations at the minimiser of the viscosity solution, without requiring nonlinear updates or full state reconstruction.

Related papers

Quantum algorithms for viscosity solutions to nonlinear Hamilton–Jacobi equations based on an entropy penalization method

Shi Jin, Nana Liu

We present a framework for efficient extraction of the viscosity solutions of nonlinear Hamilton-Jacobi equations with convex Hamiltonians. These viscosity solutions play a central role in areas such as front propagation, mean-field games, optimal control, machine learning, and a direct application to the forced Burgers' equation. Our method is based on an entropy penalization method which generalizes the Cole-Hopf transform from quadratic to general convex Hamiltonians, allowing an approximation of viscous Hamilton-Jacobi dynamics by a discrete-time linear dynamics which approximates a linear heat-like parabolic equation, and thus can also extend to continuous-time dynamics. This makes the method suitable for quantum simulation. The validity of these results hold for arbitrary nonlinearity that correspond to convex Hamiltonians, and for arbitrarily long times, thus obviating a chief obstacle in most quantum algorithms for nonlinear partial differential equations. We provide quantum algorithms-both analog and digital-for extracting pointwise values, gradients, minima, and function evaluations at the minimizer of the viscosity solution, without requiring nonlinear updates or full state reconstruction.

Proceedings of the National Academy of Sciences•Jun 10, 2026•0 citations

From Nonlinear to Unitary: Selection-Correct Linear Embeddings for Viscosity Solutions

Peng Qi

Hamilton-Jacobi equations and scalar conservation laws develop shocks and caustics in finite time, beyond which physically relevant solutions are selected by viscosity and entropy conditions. Since quantum computation natively implements linear unitary evolution, post-shock quantum algorithms require linear representations that respect these selection principles. This paper presents a self-contained account of two such selection-correct linear lifts. We first examine the quadratic Hamilton-Jacobi and viscous Burgers equations, where the Cole-Hopf transform maps the viscous regularization to the heat equation, yielding the correct vanishing-viscosity limit. We then address general convex Hamiltonians using entropy penalization and soft-min Hopf-Lax formulas to convert dynamic-programming minimizations into linear positive integral operators. We derive the discrete-time linear dynamics, prove soft-min error bounds, and demonstrate that discretization yields positive matrices with controllable operator norms. These linear maps are then explicitly embedded into unitaries via Halmos dilation, providing a direct route to quantum simulation compatible with Schrödingerization-based pipelines.

•Jan 9, 2026•0 citations

Schrödingerization based quantum algorithms for regularized Wasserstein proximal operators

Shi Jin, Nana Liu, Yue Yu

We develop a quantum algorithm for the regularized Wasserstein proximal operator, which is a fundamental tool in optimal transport and mean-field games. The regularization introduces a small diffusive term into the continuity equation of the Benamou-Brenier formulation, which results in a forward-backward PDE system consisting of a Fokker-Planck equation and a viscous Hamilton-Jacobi equation with a quadratic Hamiltonian. Through the Cole-Hopf transformation, both equations are converted to forward heat equations, whose coupling requires a Hadamard division to prepare the initial data for the second heat equation and a Hadamard product to recover the terminal density. We solve these heat equations via the Schrödingerization method and implement the Hadamard division and product operations using simple matrix-vector multiplication representations. The complete quantum algorithm prepares an $\varepsilon$-approximation of the terminal density state with $\mathcal{O}(d N_x T \log^2(1/\varepsilon))$ query complexity, up to constants depending on the potential and initial density, where $d$ is the spatial dimension, $N_x$ is the number of grid points per spatial dimension and $T$ is the evolution time. The complexity depends only {\it linearly} on $d N_x$, yielding an {\it exponential} speedup over classical methods, whose cost scales as $N_x^d$ per time step. Numerical experiments validate the effectiveness of the proposed algorithm.

arXiv:2606.28752•Jun 27, 2026•0 citations

Quantum mechanical framework for quantization-based optimization: from Gradient flow to Schroedinger equation

Jinwuk Seok, Changsik Cho

This work presents a quantum mechanical framework for analyzing quantization-based optimization algorithms. The sampling process of the quantization-based search is modeled as a gradient-flow dissipative system, leading to a Hamilton-Jacobi-Bellman (HJB) representation. Through a suitable transformation of the objective function, this formulation yields the Schroedinger equation, which reveals that quantum tunneling enables escape from local minima and guarantees access to the global optimum. By establishing the connection to the Fokker-Planck equation, the framework provides a thermodynamic interpretation of global convergence. Such an analysis between the thermodynamic and the quantum dynamic methodology unifies combinatorial and continuous optimization, and extends naturally to machine learning tasks such as image classification. Numerical experiments demonstrate that quantization-based optimization consistently outperforms conventional algorithms across both combinatorial problems and nonconvex continuous functions.

quant-pharXiv:2603.11536•Mar 13, 2026•0 citations

Quantum algorithms to simulate quadratic classical Hamiltonians and optimal control

Hari Krovi

Simulation of realistic classical mechanical systems is of great importance to many areas of engineering such as robotics, dynamics of rotating machinery and control theory. In this work, we develop quantum algorithms to estimate quantities of interest such as the kinetic energy in a given classical mechanical system in the presence of friction or damping as well as forcing or source terms, which makes the algorithm of practical interest. We show that for such systems, the quantum algorithm scales polynomially with the logarithm of the dimension of the system. We cast this problem in terms of Hamilton's equations of motion (equivalent to the first variation of the Lagrangian) and solve them using quantum algorithms for differential equations. We then consider the hardness of estimating the kinetic energy of a damped coupled oscillator system. We show that estimating the kinetic energy at a given time of this system to within additive precision is BQP hard when the strength of the damping term is bounded by an inverse polynomial in the number of qubits. We then consider the problem of designing optimal control of classical systems, which can be cast as the second variation of the Lagrangian. In this direction, we first consider the Riccati equation, which is a nonlinear differential equation ubiquitous in control theory. We give an efficient quantum algorithm to solve the Riccati differential equation well into the nonlinear regime. To our knowledge, this is the first example of any nonlinear differential equation that can be solved when the strength of the nonlinearity is asymptotically greater than the amount of dissipation. We then show how to use this algorithm to solve the linear quadratic regulator problem, which is an example of the Hamilton-Jacobi-Bellman equation.

arXiv:2404.07303•Apr 10, 2024•0 citations

Quantum algorithm for differential equations via permutation matrix representation with application to the Burgers equation

Hriday Sabharwal, Amir Kalev, Itay Hen

We develop a quantum algorithm for solving the dynamics of the nonlinear viscous Burgers equation. We apply the Carleman linearization procedure on the spatially discretized equation, followed by a padding scheme that allows implementation on qubit registers. Existing Carleman-based quantum algorithms commonly formulate the lifted linear differential equation in an oracle model. Here we decompose the padded generator into diagonal masks and reversible arithmetic permutations using the Permutation Matrix Representation (PMR), which we show to be naturally compatible with the Linear Combination of Hamiltonian Simulations (LCHS) algorithm. Under the assumptions required by LCHS - most importantly positive semidefiniteness of the Hermitian part of the linear generator, possibly after a stabilizing shift - the algorithm prepares a normalized quantum state proportional to the solution of the truncated lifted system; the stabilizing shift introduces an exponential postselection overhead, which we quantify and mitigate through a rescaling scheme. We show that our algorithm scales with the off-diagonal norm of the Carleman generator instead of the matrix norm, which can be advantageous for other generators that are diagonally dominant. We also extend the PMR scheme to general fluid equations that may contain higher-order derivatives or nonlinear terms, or may involve multiple fluid variables or spatial dimensions. The construction illustrates how PMR can serve as a convenient Hamiltonian-simulation primitive for a broader class of LCU-based algorithms.

quant-pharXiv:2608.19508•Aug 21, 2026•0 citations
We present a framework for efficient extraction of the viscosity solutions of nonlinear Hamilton-Jacobi equations with convex Hamiltonians. These viscosity solutions play a central role in areas such as front propagation, mean-field games, optimal control, machine learning, and a direct application to the forced Burgers' equation. Our method is based on an entropy penalisation method proposed by Gomes and Valdinoci, which generalises the Cole-Hopf transform from quadratic to general convex Hamiltonians, allowing a reformulation of viscous Hamilton-Jacobi dynamics by a discrete-time linear dynamics which approximates a linear heat-like parabolic equation, and can also extend to continuous-time dynamics. This makes the method suitable for quantum simulation. The validity of these results hold for arbitrary nonlinearity that correspond to convex Hamiltonians, and for arbitrarily long times, thus obviating a chief obstacle in most quantum algorithms for nonlinear partial differential equations. We provide quantum algorithms, both analog and digital, for extracting pointwise values, gradients, minima, and function evaluations at the minimiser of the viscosity solution, without requiring nonlinear updates or full state reconstruction.
We present a framework for efficient extraction of the viscosity solutions of nonlinear Hamilton-Jacobi equations with convex Hamiltonians. These viscosity solutions play a central role in areas such as front propagation, mean-field games, optimal control, machine learning, and a direct application to the forced Burgers' equation. Our method is based on an entropy penalization method which generalizes the Cole-Hopf transform from quadratic to general convex Hamiltonians, allowing an approximation of viscous Hamilton-Jacobi dynamics by a discrete-time linear dynamics which approximates a linear heat-like parabolic equation, and thus can also extend to continuous-time dynamics. This makes the method suitable for quantum simulation. The validity of these results hold for arbitrary nonlinearity that correspond to convex Hamiltonians, and for arbitrarily long times, thus obviating a chief obstacle in most quantum algorithms for nonlinear partial differential equations. We provide quantum algorithms-both analog and digital-for extracting pointwise values, gradients, minima, and function evaluations at the minimizer of the viscosity solution, without requiring nonlinear updates or full state reconstruction.
Hamilton-Jacobi equations and scalar conservation laws develop shocks and caustics in finite time, beyond which physically relevant solutions are selected by viscosity and entropy conditions. Since quantum computation natively implements linear unitary evolution, post-shock quantum algorithms require linear representations that respect these selection principles. This paper presents a self-contained account of two such selection-correct linear lifts. We first examine the quadratic Hamilton-Jacobi and viscous Burgers equations, where the Cole-Hopf transform maps the viscous regularization to the heat equation, yielding the correct vanishing-viscosity limit. We then address general convex Hamiltonians using entropy penalization and soft-min Hopf-Lax formulas to convert dynamic-programming minimizations into linear positive integral operators. We derive the discrete-time linear dynamics, prove soft-min error bounds, and demonstrate that discretization yields positive matrices with controllable operator norms. These linear maps are then explicitly embedded into unitaries via Halmos dilation, providing a direct route to quantum simulation compatible with Schrödingerization-based pipelines.
We develop a quantum algorithm for the regularized Wasserstein proximal operator, which is a fundamental tool in optimal transport and mean-field games. The regularization introduces a small diffusive term into the continuity equation of the Benamou-Brenier formulation, which results in a forward-backward PDE system consisting of a Fokker-Planck equation and a viscous Hamilton-Jacobi equation with a quadratic Hamiltonian. Through the Cole-Hopf transformation, both equations are converted to forward heat equations, whose coupling requires a Hadamard division to prepare the initial data for the second heat equation and a Hadamard product to recover the terminal density. We solve these heat equations via the Schrödingerization method and implement the Hadamard division and product operations using simple matrix-vector multiplication representations. The complete quantum algorithm prepares an ε\varepsilonε-approximation of the terminal density state with O(dNxTlog⁡2(1/ε))\mathcal{O}(d N_x T \log^2(1/\varepsilon))O(dNx​Tlog2(1/ε)) query complexity, up to constants depending on the potential and initial density, where ddd is the spatial dimension, NxN_xNx​ is the number of grid points per spatial dimension and TTT is the evolution time. The complexity depends only {\it linearly} on dNxd N_xdNx​, yielding an {\it exponential} speedup over classical methods, whose cost scales as NxdN_x^dNxd​ per time step. Numerical experiments validate the effectiveness of the proposed algorithm.
This work presents a quantum mechanical framework for analyzing quantization-based optimization algorithms. The sampling process of the quantization-based search is modeled as a gradient-flow dissipative system, leading to a Hamilton-Jacobi-Bellman (HJB) representation. Through a suitable transformation of the objective function, this formulation yields the Schroedinger equation, which reveals that quantum tunneling enables escape from local minima and guarantees access to the global optimum. By establishing the connection to the Fokker-Planck equation, the framework provides a thermodynamic interpretation of global convergence. Such an analysis between the thermodynamic and the quantum dynamic methodology unifies combinatorial and continuous optimization, and extends naturally to machine learning tasks such as image classification. Numerical experiments demonstrate that quantization-based optimization consistently outperforms conventional algorithms across both combinatorial problems and nonconvex continuous functions.
Simulation of realistic classical mechanical systems is of great importance to many areas of engineering such as robotics, dynamics of rotating machinery and control theory. In this work, we develop quantum algorithms to estimate quantities of interest such as the kinetic energy in a given classical mechanical system in the presence of friction or damping as well as forcing or source terms, which makes the algorithm of practical interest. We show that for such systems, the quantum algorithm scales polynomially with the logarithm of the dimension of the system. We cast this problem in terms of Hamilton's equations of motion (equivalent to the first variation of the Lagrangian) and solve them using quantum algorithms for differential equations. We then consider the hardness of estimating the kinetic energy of a damped coupled oscillator system. We show that estimating the kinetic energy at a given time of this system to within additive precision is BQP hard when the strength of the damping term is bounded by an inverse polynomial in the number of qubits. We then consider the problem of designing optimal control of classical systems, which can be cast as the second variation of the Lagrangian. In this direction, we first consider the Riccati equation, which is a nonlinear differential equation ubiquitous in control theory. We give an efficient quantum algorithm to solve the Riccati differential equation well into the nonlinear regime. To our knowledge, this is the first example of any nonlinear differential equation that can be solved when the strength of the nonlinearity is asymptotically greater than the amount of dissipation. We then show how to use this algorithm to solve the linear quadratic regulator problem, which is an example of the Hamilton-Jacobi-Bellman equation.
We develop a quantum algorithm for solving the dynamics of the nonlinear viscous Burgers equation. We apply the Carleman linearization procedure on the spatially discretized equation, followed by a padding scheme that allows implementation on qubit registers. Existing Carleman-based quantum algorithms commonly formulate the lifted linear differential equation in an oracle model. Here we decompose the padded generator into diagonal masks and reversible arithmetic permutations using the Permutation Matrix Representation (PMR), which we show to be naturally compatible with the Linear Combination of Hamiltonian Simulations (LCHS) algorithm. Under the assumptions required by LCHS - most importantly positive semidefiniteness of the Hermitian part of the linear generator, possibly after a stabilizing shift - the algorithm prepares a normalized quantum state proportional to the solution of the truncated lifted system; the stabilizing shift introduces an exponential postselection overhead, which we quantify and mitigate through a rescaling scheme. We show that our algorithm scales with the off-diagonal norm of the Carleman generator instead of the matrix norm, which can be advantageous for other generators that are diagonally dominant. We also extend the PMR scheme to general fluid equations that may contain higher-order derivatives or nonlinear terms, or may involve multiple fluid variables or spatial dimensions. The construction illustrates how PMR can serve as a convenient Hamiltonian-simulation primitive for a broader class of LCU-based algorithms.