PIDM-DP: Physics-Informed Diffusion with Dormand-Prince Integration for Chaotic System Identification and State Reconstruction across Multiple Dynamical Regimes

Shailendra Dabral · 2026

Read PIDM-DP: Physics-Informed Diffusion with Dormand-Prince Integration for Chaotic System Identification and State Reconstruction across Multiple Dynamical Regimes by Shailendra Dabral (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2605-26619.txt.

Introduction

Shailendra Dabral rebuilds chaotic trajectories from sparse, noisy peeks with a physics-guided diffusion model called PIDM-DP. A fifth-order Dormand-Prince solver sits inside the denoising loop, so each generated path must obey the governing equations. A ramped physics weight stops gradient blowups on stiff systems. Across five benchmarks at ten percent observation density, it improves reconstruction error up to 15.4-fold over plain diffusion and beats an ensemble Kalman filter on stiff cases, while preserving chaotic invariants.

Catalog record

CREATORShailendra Dabral
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE57767
LICENSEhttp://creativecommons.org/licenses/by-sa/4.0/
RIGHTSCC BY SA
SOURCEarXiv

Source: https://arxiv.org/abs/2605.26619

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

PIDM-DP: Physics-Informed Diffusion with Dormand-Prince Integration for Chaotic System Identification and State Reconstruction across Multiple Dynamical Regimes Shailendra Dabral Abstract Reconstructing continuous state trajectories of chaotic dynamical systems from sparse, noisy observations remains a fundamental open problem in nonlinear science. We introduce the Physics-Informed Diffusion Model with Dormand-Prince Integration (PIDM-DP) , which embeds a fully differentiable 5th-order Dormand-Prince (DP-RK45) ODE integrator directly into the reverse sampling loop of a Denoising Diffusion Probabilistic Model (DDPM). At each denoising step, physics residuals are back-propagated via automatic differentiation, constraining every generated trajectory to satisfy the system’s governing equations to 5th-order accuracy. A linear-scheduled guidance mechanism that ramps the physics weight from zero at high noise levels to its full value near the clean-data limit prevents the gradient explosions that cause naive physics-informed approaches to fail on stiff systems with Jacobian eigenvalues of order O ⁡ ( 10 3 ) O(10^{3}) . Evaluated across five benchmark systems of increasing complexity 3D Lorenz, 3D Rössler, 5D Hyperchaotic, 20D Lorenz-96, and the stiff 3D Rabinovich-Fabrikant at 10% observation density with additive Gaussian noise ( σ = 0.05 \sigma{=}0.05 ), PIDM-DP achieves reconstruction RMSE improvements of up to 15.4 × 15.4\times over an unconstrained diffusion baseline and decisively outperforms the Ensemble Kalman Filter on stiff systems where ensemble covariance collapses. On the Rabinovich-Fabrikant out-of-distribution benchmark, PIDM-DP attains RMSE 0.1097 ± 0.0269 \mathbf{0.1097\pm 0.0269} versus 0.9443 ± 0.5288 0.9443\pm 0.5288 (unconstrained diffusion, 8.6 × 8.6\times worse) and 0.3561 ± 0.3040 0.3561\pm 0.3040 (EnKF, 3.2 × 3.2\times worse), with p < 0.001 p{<}0.001 in paired Wilcoxon tests ( N = 30 N{=}30 ). Topological validation via the Rosenstein Lyapunov estimator confirms that PIDM-DP preserves the chaotic invariant measure. The joint state-parameter representation enables implicit system identification, recovering the Rayleigh number with 5.3 % 5.3\% mean absolute percentage error from 10% observations. PIDM-DP achieves the lowest mean RMSE against CSDI, GRU-ODE, and ESN baselines across all 10 in-distribution and out-of-distribution benchmarks. Contributions. PIDM-DP embeds a differentiable DP-RK45 integrator into reverse diffusion sampling; linear-scheduled physics guidance prevents gradient explosions on stiff chaotic systems; on the Rabinovich-Fabrikant OOD benchmark the method improves RMSE by 8.6\times over unconstrained diffusion and 3.2\times over EnKF; it achieves the best mean RMSE across all 10 ID/OOD benchmarks against CSDI, GRU-ODE, and ESN; and Lyapunov plus parameter-recovery analyses support manifold-level fidelity. 1 Introduction 1.1 Motivation and Problem Setting The problem of reconstructing a continuous state trajectory from a sparse, noisy measurement sequence is among the most practically consequential inverse problems in nonlinear science. In atmospheric physics, satellite overpasses deliver state measurements at intervals of approximately 90 minutes, each corrupted by instrument noise; a complete atmospheric state must be inferred from this sliver of information to initialise operational 72-hour forecasts [ 19 ] . In plasma physics, sparse magnetic probe arrays must reconstruct the full current distribution for real-time tokamak control [ 22 ] . In neuroscience, a handful of electrode recordings must capture whole-brain activity from spatially localised measurements [ 10 ] . In each setting the same mathematical challenge recurs: given m\ll L noisy observations \{y_{1},\ldots,y_{m}\} sampled from an L -step trajectory of a chaotic ODE, reconstruct the full continuous state \mathbf{x}(t) and identify the hidden system parameters \mathbf{p} . What distinguishes this from a standard regression or interpolation problem is the butterfly effect [ 31 ] . A 1% error in the reconstructed state at one observation grows by a factor of e^{\lambda_{\max}\Delta t_{\rm gap}} before the next observation arrives. For the Lorenz system with \lambda_{\max}\approx 0.906 and 90 unobserved steps at \Delta t=0.05 , this amplification exceeds 59-fold. Any method that does not actively enforce the governing ODE will generate trajectories that are statistically plausible but physically impossible. 1.2 Limitations of Existing Approaches Ensemble Kalman Filter (EnKF). The EnKF [ 17 , 8 , 9 ] propagates an ensemble of states forward under the exact governing equations and applies Kalman corrections at each observation. As the operational gold standard for atmospheric data assimilation, it performs optimally under near-Gaussian, mildly nonlinear conditions. However, at extreme observation sparsity (90% missing data), three systematic limitations emerge: (i) the Gaussian error assumption distorts the fractal geometry of strange attractors; (ii) ensemble collapse occurs when the effective sample size is too small relative to the state dimension [ 14 ] ; and (iii) for stiff systems such as the Rabinovich-Fabrikant equations, ensemble members entering the stiff manifold during unobserved integration windows generate covariance explosions that standard inflation techniques cannot recover [ 1 ] . Modern sequence and generative models, LSTMs, Echo State Networks, Transformers, Neural ODEs, and diffusion-based imputers [ 13 , 16 , 34 , 3 , 33 ] are powerful statistical interpolators that learn correlations rather than physics. As we demonstrate empirically across five benchmarks, a diffusion model trained on chaotic trajectories without physical constraints undergoes Lyapunov collapse : the maximal Lyapunov exponent of generated trajectories drops toward zero, indicating that the model has learned smooth, nearly-periodic orbits rather than genuine strange attractors. Physics-Informed Neural Networks (PINNs). PINNs [ 23 , 18 ] incorporate ODE residuals into the training loss but require per-trajectory optimisation, cannot provide amortised inference, and typically employ low-order numerical residuals that accumulate significant error on chaotic trajectories. Crucially, they provide no mechanism for applying physics constraints selectively during different phases of generative sampling. 1.3 The Present Contribution We introduce PIDM-DP (Physics-Informed Diffusion Model with Dormand-Prince integration), which resolves all three failure modes through three tightly integrated innovations: 1. Differentiable DP-RK45 residual. A PyTorch-native six-stage Runge-Kutta implementation with explicit rational Butcher coefficients provides a 5th-order accurate physics gradient at every reverse diffusion step. The implementation achieves <10^{-14} absolute error against a NumPy reference integrator, validated before all experiments. 2. Linear-scheduled guidance. A time-varying physics weight \lambda_{\rm phy}(t)=\lambda_{\rm base}(1-t/T) suppresses physics corrections during the high-noise phase where off-manifold states produce meaningless and numerically destructive gradients, and applies the full constraint progressively as the sample converges toward the clean-data limit. This scheduling mechanism is, to our knowledge, the first solution to the stiffness-gradient instability that causes existing physics-informed diffusion approaches to fail on systems with Jacobian spectral radii of O(10^{2}) – O(10^{3}) . 3. Safe autograd manifold projection. Gradient norm clipping combined with a graceful exception-handling fallback ensures the reverse diffusion degrades to standard DDPM rather than diverging when the integrator encounters a numerically singular state. The framework further incorporates a joint state-parameter representation that co-generates trajectory reconstructions and hidden system parameter estimates in a single forward pass, without any direct parameter observations or per-trajectory retraining. We evaluate PIDM-DP across five chaotic benchmarks at 10% observation density: the 3D Lorenz, 3D Rössler, 5D Hyperchaotic, 20D Lorenz-96, and the notoriously stiff 3D Rabinovich-Fabrikant systems. PIDM-DP achieves the lowest mean RMSE in all 10 in-distribution (ID) and out-of-distribution (OOD) benchmark conditions against CSDI, GRU-ODE, and ESN baselines, with 29 of 30 paired Wilcoxon tests significant at p<0.05 . On the stiff Rabinovich-Fabrikant system, PIDM-DP outperforms both the unconstrained diffusion baseline and the EnKF oracle by factors of 8.6\times and 3.2\times , respectively. 1.4 Paper Organisation Section 2 reviews chaotic dynamics, DDPMs, the Dormand-Prince method, and the EnKF, together with a discussion of related work. Section 3 introduces the five benchmark systems. Section 4 presents the PIDM-DP architecture and

Read the complete text