Stochastic Port-Hamiltonian Neural Networks: Universal Approximation with Passivity Guarantees

• URL: http://arxiv.org/abs/2603.10078v1
• Date: Tue, 10 Mar 2026 09:16:15 GMT
• Title: Stochastic Port-Hamiltonian Neural Networks: Universal Approximation with Passivity Guarantees
• Abstract summary: We introduce port-Hamiltonian neural networks, which parameterize the Hamiltonian with a feedforward network.<n>Experiments show improved long horizon rollouts and reduced energy error relative to a multilayer perceptron baseline.
• Score: 0.688204255655161
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Stochastic port-Hamiltonian systems represent open dynamical systems with dissipation, inputs, and stochastic forcing in an energy based form. We introduce stochastic port-Hamiltonian neural networks, SPH-NNs, which parameterize the Hamiltonian with a feedforward network and enforce skew symmetry of the interconnection matrix and positive semidefiniteness of the dissipation matrix. For Itô dynamics we establish a weak passivity inequality in expectation under an explicit generator condition, stated for a stopped process on a compact set. We also prove a universal approximation result showing that, on any compact set and finite horizon, SPH-NNs approximate the coefficients of a target stochastic port-Hamiltonian system with $C^2$ accuracy of the Hamiltonian and yield coupled solutions that remain close in mean square up to the exit time. Experiments on noisy mass spring, Duffing, and Van der Pol oscillators show improved long horizon rollouts and reduced energy error relative to a multilayer perceptron baseline.

Related papers

• Interpretable Activation-Selection Neural Networks for Symbolic Regression of Parameter-Dependent Hamiltonian Eigenvalues [45.88028371034407]
  We introduce an activation-selection network (ASN), a differentiable symbolic-regression architecture.<n>Each input node learns a sparse combination of predefined analytical functions, and the trained network can be converted directly into an explicit expression.<n>We apply the ASN to effective three- and four-site spin-chain Hamiltonians relevant to zero-quantum nuclear magnetic resonance.
  arXiv  Detail & Related papers  (2026-08-26T07:45:10Z)
• Quantum Port-Hamiltonian Neural Networks: Learning Conservative and Dissipative Dynamics via Measurement-Induced Nonlinearity [0.09229852843814058]
  We introduce Quantum Port-Hamiltonian Neural Networks (Q-pHNNs)<n>Q-pHNNs learn classical dynamics in a structure-preserving manner.
  arXiv  Detail & Related papers  (2026-07-14T02:16:20Z)
• Hamiltonian dynamics from pure dissipation [3.1798318618973362]
  We show internal Hamiltonian dynamics can be "faked" via external pure dissipation.<n>In a GKSL representation with zero explicit Hamiltonian term but nontraceless jump operators, dissipative generators can approximate Hamiltonian dynamics within $$ error.
  arXiv  Detail & Related papers  (2026-04-20T17:28:28Z)
• Continuum field theory of matchgate tensor network ensembles [36.94429692322632]
  We develop a continuum description for random ensembles of two-dimensional fermionic matchgate tensor networks.<n>As a diagnostic of the resulting universal physics, we analyze disorder-averaged moments of fermionic two-point functions.<n>We show that disorder drives universal long-distance behavior governed by a nonlinear sigma-model of symmetry class D with a topological term.
  arXiv  Detail & Related papers  (2026-03-06T12:15:35Z)
• Learning Passive Continuous-Time Dynamics with Multistep Port-Hamiltonian Gaussian Processes [0.0]
  We propose the multistep port-Hamiltonian Gaussian process (MS-PHS GP) to learn physically consistent continuous-time dynamics and a posterior over the Hamiltonian from noisy, irregularly-sampled trajectories.<n>MS-PHS GP enables closed-form conditioning of both the vector field and the Hamiltonian surface without latent states, while enforcing energy balance and passivity by design.<n>We demonstrate improved vector-field recovery and well-calibrated Hamiltonian uncertainty on mass-spring, Van der Pol, and Duffing benchmarks.
  arXiv  Detail & Related papers  (2025-10-01T00:55:28Z)
• Learning Generalized Hamiltonian Dynamics with Stability from Noisy Trajectory Data [4.680663665263333]
  We introduce a robust framework for learning various generalized Hamiltonian dynamics from noisy, sparse phase-space data.<n>It is challenging for a single Hamiltonian network model to capture the distinctive and varying motion dynamics and physics of a phase space.
  arXiv  Detail & Related papers  (2025-09-08T23:29:04Z)
• The Hamiltonian of Poly-matrix Zero-sum Games [0.276240219662896]
  We identify the Hamiltonian function that generates the dynamics of poly-matrix zero-sum games.<n>We reveal the symmetries of our Hamiltonian and derive the associated conserved quantities.<n>Our results highlight the potential of Hamiltonian dynamics in uncovering the structural properties of learning dynamics in games.
  arXiv  Detail & Related papers  (2025-05-19T01:46:29Z)
• Weak coupling limit for quantum systems with unbounded weakly commuting system operators [50.24983453990065]
  This work is devoted to a rigorous analysis of the weak coupling limit (WCL) for the reduced dynamics of an open infinite-dimensional quantum system interacting with electromagnetic field or a reservoir formed by Fermi or Bose particles.<n>We derive in the weak coupling limit the reservoir statistics, which is determined by whose terms in the multi-point correlation functions of the reservoir are non-zero in the WCL.<n>We prove that the resulting reduced system dynamics converges to unitary dynamics with a modified Hamiltonian which can be interpreted as a Lamb shift to the original Hamiltonian.
  arXiv  Detail & Related papers  (2025-05-13T05:32:34Z)
• Quantum Simulation of Nonlinear Dynamical Systems Using Repeated Measurement [42.896772730859645]
  We present a quantum algorithm based on repeated measurement to solve initial-value problems for nonlinear ordinary differential equations. We apply this approach to the classic logistic and Lorenz systems in both integrable and chaotic regimes.
  arXiv  Detail & Related papers  (2024-10-04T18:06:12Z)
• Fourier Neural Operators for Learning Dynamics in Quantum Spin Systems [77.88054335119074]
  We use FNOs to model the evolution of random quantum spin systems. We apply FNOs to a compact set of Hamiltonian observables instead of the entire $2n$ quantum wavefunction.
  arXiv  Detail & Related papers  (2024-09-05T07:18:09Z)
• Symmetric Mean-field Langevin Dynamics for Distributional Minimax Problems [78.96969465641024]
  We extend mean-field Langevin dynamics to minimax optimization over probability distributions for the first time with symmetric and provably convergent updates. We also study time and particle discretization regimes and prove a new uniform-in-time propagation of chaos result.
  arXiv  Detail & Related papers  (2023-12-02T13:01:29Z)
• Stoquasticity in circuit QED [78.980148137396]
  We show that scalable sign-problem free path integral Monte Carlo simulations can typically be performed for such systems. We corroborate the recent finding that an effective, non-stoquastic qubit Hamiltonian can emerge in a system of capacitively coupled flux qubits.
  arXiv  Detail & Related papers  (2020-11-02T16:41:28Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.