Unifying Physical Backpropagation

• URL: http://arxiv.org/abs/2608.11585v1
• Date: Wed, 12 Aug 2026 02:53:39 GMT
• Title: Unifying Physical Backpropagation
• Abstract summary: Physical computing systems exploit device dynamics for computation, but their gradient-based optimization is challenging.<n>We propose a unifying theory identifying when a physical system can compute the gradient of its own performance.<n>We show that reciprocity is only the simplest instance of a more general intertwining condition, which extends exact on-device gradient computation to a class of non-Hermitian, non-reciprocal systems.
• Score: 0.0
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Physical computing systems exploit device dynamics for computation, but their gradient-based optimization is challenging: backpropagation through a digital twin suffers from model-reality gap. On-device gradient computation could resolve this issue, and a handful of theoretical and experimental studies have proposed ways to achieve it. Yet a unifying theory identifying when a physical system can compute the gradient of its own performance has been missing. Here we develop such a unification, based on the adjoint method: we identify sufficient conditions under which the adjoint field required for formally exact gradients can be generated on the same hardware that performs the computation. Linear and nonlinear systems obey fundamentally different conditions: for linear systems damping or gain is admissible provided reciprocity is preserved. For nonlinear trajectory systems the sufficient conditions are reciprocity of the linearized system and the existence of a time-reversal mirror. Algorithmically, the nonlinear case requires infinitesimal nudging, whereas linear systems admit a finite-amplitude experiment. We recover Equilibrium Propagation, Hamiltonian echo backpropagation, fully forward mode training and in situ gradient methods in integrated-photonic and free-space-optical systems. We further show that reciprocity is only the simplest instance of a more general intertwining condition, which extends exact on-device gradient computation to a class of non-Hermitian, non-reciprocal systems. Further generalizations include time-dependent parameters, Onsager-reciprocal dynamics and nonlinear, PT-symmetric Schrödinger equations. Our work provides a unified theoretical basis for formally exact physical learning algorithms and a template for constructing them across a range of physical systems.

Related papers

• Alleviating Post-Linearization Challenges for Solving Nonlinear Systems on a Quantum Computer [0.0]
  Carleman linearization provides a high dimensional infinite linear system corresponding to a finite nonlinear system.<n>We decompose the Hamiltonian into the weighted sum of non-unitary operators, namely the Sigma basis.<n>Once the Hamiltonian is decomposed, we then use the concept of unitary completion to construct the circuit.
  arXiv  Detail & Related papers  (2026-02-06T14:56:39Z)
• Disordered Dynamics in High Dimensions: Connections to Random Matrices and Machine Learning [52.26396748560348]
  We provide an overview of high dimensional dynamical systems driven by random matrices.<n>We focus on applications to simple models of learning and generalization in machine learning theory.
  arXiv  Detail & Related papers  (2026-01-03T00:12:32Z)
• Identifiability Challenges in Sparse Linear Ordinary Differential Equations [4.895067344504143]
  We show that sparse systems are unidentifiable with a positive probability in practically relevant sparsity regimes.<n>We further study empirically how this theoretical unidentifiability manifests in state-of-the-art methods to estimate linear ODEs from data.
  arXiv  Detail & Related papers  (2025-06-11T14:55:36Z)
• Learning Nonlinear Dynamics in Physical Modelling Synthesis using Neural Ordinary Differential Equations [13.755383470312001]
  A modal decomposition leads to a coupled nonlinear system of ordinary differential equations.<n>Recent work in applied machine learning approaches has been used to model lumped dynamic systems automatically from data.<n>We show that the model can be trained to reproduce the nonlinear dynamics of the system.
  arXiv  Detail & Related papers  (2025-05-15T17:17:21Z)
• Identifiability and Asymptotics in Learning Homogeneous Linear ODE Systems from Discrete Observations [114.17826109037048]
  Ordinary Differential Equations (ODEs) have recently gained a lot of attention in machine learning. theoretical aspects, e.g., identifiability and properties of statistical estimation are still obscure. This paper derives a sufficient condition for the identifiability of homogeneous linear ODE systems from a sequence of equally-spaced error-free observations sampled from a single trajectory.
  arXiv  Detail & Related papers  (2022-10-12T06:46:38Z)
• Agnostic Physics-Driven Deep Learning [82.89993762912795]
  This work establishes that a physical system can perform statistical gradient learning without gradient computations. In Aeqprop, the specifics of the system do not have to be known: the procedure is based on external manipulations. Aeqprop also establishes that in natural (bio)physical systems, genuine gradient-based statistical learning may result from generic, relatively simple mechanisms.
  arXiv  Detail & Related papers  (2022-05-30T12:02:53Z)
• Challenges for quantum computation of nonlinear dynamical systems using linear representations [2.2000635322691378]
  We show that a necessary projection into a feasible finite-dimensional space will in practice induce numerical artifacts which can be hard to eliminate or even control. As a result, a practical, reliable and accurate way to use quantum computation for solving general nonlinear dynamical systems is still an open problem.
  arXiv  Detail & Related papers  (2022-02-04T15:29:53Z)
• Exact solutions of interacting dissipative systems via weak symmetries [77.34726150561087]
  We analytically diagonalize the Liouvillian of a class Markovian dissipative systems with arbitrary strong interactions or nonlinearity. This enables an exact description of the full dynamics and dissipative spectrum. Our method is applicable to a variety of other systems, and could provide a powerful new tool for the study of complex driven-dissipative quantum systems.
  arXiv  Detail & Related papers  (2021-09-27T17:45:42Z)
• On the static effective Hamiltonian of a rapidly driven nonlinear system [0.0]
  We present a formula for the computation of the static effective Hamiltonian of a system under a fast-oscillating drive. Our result is well-suited to symbolic calculations performed by a computer.
  arXiv  Detail & Related papers  (2021-08-05T21:39:22Z)
• Linear embedding of nonlinear dynamical systems and prospects for efficient quantum algorithms [74.17312533172291]
  We describe a method for mapping any finite nonlinear dynamical system to an infinite linear dynamical system (embedding) We then explore an approach for approximating the resulting infinite linear system with finite linear systems (truncation)
  arXiv  Detail & Related papers  (2020-12-12T00:01:10Z)
• On dissipative symplectic integration with applications to gradient-based optimization [77.34726150561087]
  We propose a geometric framework in which discretizations can be realized systematically. We show that a generalization of symplectic to nonconservative and in particular dissipative Hamiltonian systems is able to preserve rates of convergence up to a controlled error.
  arXiv  Detail & Related papers  (2020-04-15T00:36:49Z)

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.