Pith. sign in

REVIEW 3 major objections 7 minor 62 references

rodeo: Probabilistic Methods of Parameter Inference for Ordinary Differential Equations

T0 review · 3 major / 7 minor · reviewed 2026-08-06 · deepseek-v4-flash

Pith's one-line read rodeo, a JAX-based Python library, claims that probabilistic ODE parameter inference can scale linearly in both evaluation points and system variables, matching deterministic-solver accuracy while running 2–12x faster in its tested…

desk verdict A useful JAX/Python library paper that deserves peer review, provided the authors tighten the speed claims, add a Jacobian ablation, and quantify 'indistinguishable'. read the letter →

arxiv 2506.21776 v1 pith:EX3SOIIJ submitted 2025-06-26 stat.CO

classification stat.CO
keywords probabilisticODEsolversparameterinferenceBayesianfilteringKalmanfilterJAXintegratedBrownianmotionblock-diagonalcomputationdynamicalsystems
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper presents rodeo, a Python library that makes probabilistic ODE solvers practical for parameter estimation. Its central claim is that solving an ODE with quantified numerical uncertainty—by treating the solver as a Bayesian filter—can be done in time linear in both the number of grid points and the number of system variables, and that the resulting parameter posteriors match those of a high-accuracy deterministic solver at a fraction of the compute. Correct, this would let practitioners report uncertainty from numerical integration as part of their parameter estimates without paying the cost usually associated with probabilistic methods. The library unifies five inference approaches (Basic, Fenrir, DALTON, marginal MCMC, and a Markov-process version of MAGI) under one interface, supporting partially observed systems and non-Gaussian measurement models.

What carries the argument

The central object is the Kalman-filter probabilistic ODE solver with block-diagonal 'model interrogation.' A Gaussian Markov process prior (by default integrated Brownian motion, giving transition matrix $Q$ and covariance $R$) is placed on the solution and its derivatives, and the ODE residual $W\mathbf{X}_n - f(\mathbf{X}_n, t_n)$ is treated as a zero-mean observation $Z_n = 0$ in the state-space model, so that posterior draws and means come from Kalman filtering and smoothing. The claim of linear-in-$d$ scaling rests on the block-diagonal structure of the weight matrix $W$ together with the modified Jacobian $J^*_f$, which zeroes out the off-diagonal blocks of the full Jacobian so that each variable's Kalman recursions run independently. That device, plus JAX's automatic differentiation and JIT compilation, is what carries the speed-accuracy results.

What would settle it

Run the same parameter-inference benchmark twice at a fixed step size—once with the Krämer block-diagonal Jacobian interrogation and once with the full Tronarp Jacobian—and measure the distance between the two resulting parameter posteriors (e.g., Wasserstein distance). If the posteriors separate by more than the solver's own uncertainty at moderate coupling strength, the 'minimal loss in accuracy' premise fails and the linear-scaling speed advantage no longer comes free.

Watch

Extended reading notes

Core claim

The paper claims that the traditional cost gap between probabilistic and deterministic ODE solvers can be closed by combining a Kalman-filter formulation of the solver with variable-wise block-diagonal structure. The ODE residual is treated as a zero-mean pseudo-observation in a nonlinear state-space model; a first-order Taylor 'model interrogation' linearizes it; and by keeping only the block-diagonal part of the resulting Jacobian (the Krämer-modified $J^*_f$), the Kalman recursions decompose per system variable, reducing complexity from $O(d^3)$ to $O(d)$ in the number of variables $d$. On top of this solver, several likelihood approximations from the literature—the plug-in Basic method, Fenrir, DALTON, the Chkrebtii marginal MCMC, and a Markov-prior version of MAGI—are expressed uniformly and implemented with JAX automatic differentiation and just-in-time compilation. In the FitzHugh-Nagumo, Hes1, and SEIRAH benchmarks, the paper shows these posteriors becoming indistinguishable from a high-accuracy RKDP deterministic solver's posterior at moderate step sizes, while running 2–12 times faster than LSODA and RKDP.

Load-bearing premise

The load-bearing premise is that zeroing out the off-diagonal blocks of the ODE Jacobian (the Krämer modification) causes minimal accuracy loss; the paper asserts 'extensive evidence' for this but reports no quantitative comparison, so for strongly coupled systems the linear-time solver's posterior could drift from the full-Jacobian answer.

Editorial extensions

If this is right

  • Parameter posteriors from the Basic, Fenrir, and DALTON approximations converge to the deterministic high-accuracy posterior as the solver grid refines, so numerical integration error can be treated as part of the statistical model instead of being hidden.
  • Because the block-diagonal smoother keeps memory linear in the number of grid points, rodeo can backpropagate directly through the solver steps, making gradient-based inference faster than adjoint-method alternatives.
  • The same machinery covers partially observed systems (Hes1 with unobserved $H(t)$, SEIRAH with two observed compartments) and non-Gaussian noise (Poisson count data), so the probabilistic-solver benefit extends beyond textbook Gaussian settings.
  • Blocking itself is worth 3–4× over the unblocked $O(d^3)$ solver in the reported timings, and the full solver is 2–12× faster than LSODA and RKDP at matched posterior accuracy.
  • As the number of evaluation points $N$ grows, the solver's posterior mean approaches the true ODE solution while its uncertainty bands shrink, meaning the uncertainty output is a usable diagnostic for under-resolved discretizations.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • A direct accuracy comparison between the block-diagonal Jacobian and the full Jacobian is the missing experiment; a natural follow-up would measure posterior divergence versus coupling strength to map where linear scaling holds.
  • If the insensitivity to the IBM prior scale noted in the examples holds generally, the solver's reported uncertainty bands are effectively determined by discretization error and the data rather than the user's prior choice—a testable calibration property.
  • The same blocking device could extend to time-varying parameters, which the paper names as future work, with no change to the $O(d)$ complexity.
  • If the speed advantage replicates on stiff systems—listed by the authors as an open direction—the practical case for replacing deterministic integrators in routine model calibration would be much stronger.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 7 minor

Summary. The paper presents rodeo, a Python library built on JAX for probabilistic parameter inference in ODEs. The library implements a Bayesian filtering framework with Gaussian Markov process priors, several model interrogation methods (Chkrebtii et al., Schober et al., Tronarp et al., Krämer et al.), and multiple inference algorithms: a basic plug-in likelihood, Fenrir, DALTON (Gaussian and non-Gaussian), marginal MCMC, and a Markov-prior variant of MAGI. The central claims are that the solver scales linearly in the number of discretization points N and system variables d (via block-diagonal interrogation and a modified Jacobian), and that inference is fast, accurate, and scalable, as demonstrated on four examples (a second-order univariate ODE, FitzHugh-Nagumo, Hes1, and SEIRAH) with comparisons to LSODA and RKDP. The paper is written primarily as a software and methods showcase, with extensive code listings and pseudocode for each algorithm.

Significance. If the claims hold, rodeo is a valuable contribution: it provides a JAX-native, AD-compatible, JIT-compiled implementation of several published probabilistic ODE inference methods in one package, making them accessible to Python users. The paper includes detailed pseudocode, runnable examples, and benchmarks against independent deterministic solvers (LSODA and RKDP via difrax), which are welcome strengths. The generalization of MAGI to arbitrary-order ODEs with Markov priors is a useful extension. However, the headline "linear scaling" and "fast" claims rest on a block-diagonal Jacobian approximation whose accuracy is asserted but not quantitatively demonstrated, and the speed summary in Section 5.13 is not consistent with Table 3. These issues are fixable but should be addressed before the paper can be recommended for publication.

major comments (3)
  1. [§2.3] The linear-in-d complexity claim depends on the block-diagonal modified Jacobian J*_f (Krämer et al. 2021), which zeroes all off-diagonal blocks of the Jacobian. The paper states that "there is extensive evidence in our experiments and those of Krämer et al. (2021) suggesting there is minimal loss in accuracy compared to using the full Jacobian," but no quantitative comparison is reported in this manuscript. This is load-bearing because the speed-accuracy trade-off in the numerical examples is obtained only under this approximation. Please add a numerical study on a strongly coupled ODE (e.g., with large off-diagonal Jacobian entries) comparing the posterior produced by J* versus the full Jacobian at fixed N, or explicitly state the limitation that the linear-scaling claim holds only when the block-diagonal approximation is accurate.
  2. [§5.13, Table 3] The text says the last three examples show rodeo is "2 to 12 times faster than RKDP and 2-12 times faster than LSODA," but the RKDP column for those rows is 3.94, 3.53, and 2.86, i.e., a range of 2.86-3.94, not 2-12. The 12.64 figure comes from the univariate Chkrebtii example (N=30), which is not in the emphasized multivariate set. Additionally, the column labeled "rodeo (no blocking)" is ambiguous: the values 3.49, 2.77, 4.42 appear to indicate the speedup factor from blocking (consistent with the text that "blocking is 3-4 times faster"), but the header suggests it reports the speed of the no-blocking solver itself. Please correct the summary ranges and clarify what the last column measures.
  3. [§5.13, Table 3] The speed comparisons are based on the smallest N for which a Laplace posterior is "indistinguishable" from the RKDP posterior. This criterion is not formalized or quantified. Without a concrete measure (e.g., Wasserstein distance between marginal posteriors, KL divergence, or a check that credible interval coverages match), the reported speedups are not reproducible and the "fast, accurate" claim is not fully supported. Please specify the comparison metric used to select N.
minor comments (7)
  1. [§2.3, Eq. (14)] In the blocking definitions, the blocks a^(k)_n, B^(k)_n, and V^(k)_n are described as having dimensions p_k×1, p_k×p_k, and p_k×p_k, respectively. However, these quantities must have the same number of rows as the ODE equations for variable k (r_k); for example, in a first-order univariate ODE with three state derivatives, B_n is 1×3, not 3×3. Please correct the notation to make the dimensions consistent with the observation model W^(k) + B^(k) having shape r_k × p_k.
  2. [§1, Table 1] The table caption says "The proposed method in the accompanying reference(s)..." but the table does not list rodeo itself; it lists only existing methods. Please clarify whether the table is intended to include rodeo or to survey prior work.
  3. [§3.5] There is a typo: "An significant contribution" should be "A significant contribution."
  4. [§3.5, Eq. (39)] The notation pβ(Θ, Ũ1:N | Z1:N, Y0:M) omits the conditioning value Z1:N = 0; the posterior should be written pβ(Θ, Ũ1:N | Z1:N = 0, Y0:M) for consistency with Equation (38).
  5. [§5.10] The statement that unobserved components are handled by setting the observation variance to zero and using "the log density of Normal(0; 0, 0)" is not mathematically well-defined, since a normal distribution with zero variance is degenerate. Please clarify how the code actually handles these entries (e.g., by masking or by using a small positive variance).
  6. [§5.7] The formula for the MAGI prior temperature β in the text (β = η^{-2} Δt^{2-2q} Δt') and the code (beta = dt_obs * dt_sim ** (2 - 2 * n_deriv) * sigma[0] ** (-2)) should be cross-checked for notational consistency, particularly the role of q versus n_deriv, so that readers can reproduce the choice.
  7. [Throughout] There are numerous typographical and formatting issues in the code listings and acknowledgments (e.g., "Cananda" for "Canada", inconsistent spacing in comments, stray characters). A careful proofreading pass would improve the presentation.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the solver and inference methods are implemented from published algorithms and benchmarked against independent deterministic solvers.

full rationale

The paper's central claims—linear scaling, speed, and accurate parameter inference—do not reduce to their own inputs by construction. The O(d) blocking is an explicitly stated design choice in Section 2.3, obtained by requiring block-diagonal W, B, V, Q, and R and then performing Kalman recursions blockwise in Algorithm 2; the linear-scaling claim is a complexity property of that disclosed algorithm, not a prediction fitted from data. The Krämer et al. (2021) modified Jacobian J*_f is imported from prior work and the paper asserts 'extensive evidence ... minimal loss in accuracy' without reporting a quantitative full-Jacobian comparison; this is an under-supported approximation and a correctness risk for strongly coupled systems, but it is not circular because the approximation is not defined in terms of the target speed or accuracy results. DALTON is self-cited (Wu and Lysy 2024), but it is a published method with its own derivation, and the numerical comparisons are anchored to independent external solvers (LSODA, RKDP, diffrax) and to closed-form or known-truth solutions. No fitted parameter is renamed as a prediction: the examples compare posterior approximations at various step sizes against RKDP posteriors and true parameter values. The limitations stated in Section 6—no adaptive step-size selection, no time-varying parameters, and no stiff-system or boundary-value tests—are scope restrictions rather than circular reasoning. Overall, the derivation chain is self-contained with respect to its benchmark claims, and no circular step can be exhibited from the paper's equations or citations.

Assumptions & free parameters 1 free parameters · 4 assumptions · 0 invented entities

The central claims rest on standard probabilistic-numerics assumptions inherited from the literature: the IBM prior, the linearized surrogate model, and the block-diagonal Jacobian heuristic. The only hand-chosen number is the IBM scale sigma=0.1, which the authors argue is insensitive. No new entities are postulated.

free parameters (1)
  • IBM prior scale sigma (also called eta in MAGI) = 0.1
    Set by hand in all examples (Sections 5.1 and 5.7). Authors state insensitivity to this choice over orders of magnitude, so it is not fitted to data, but it is a hand-chosen constant that affects the prior covariance.
assumptions (4)
  • domain assumption The ODE solution process follows a q-times integrated Brownian motion prior (Eq. 8).
    This Gaussian Markov prior is the backbone of the solver (Section 2.2). It is a modeling assumption inherited from Schober et al. (2019).
  • domain assumption The linearized surrogate model (5) obtained via model interrogation adequately approximates the intractable posterior (4).
    The Kalman solver computes the posterior under the Taylor-linearized model, not the exact nonlinear model. Convergence as N grows is cited from Kersting et al. (2020b), but for finite N the approximation error is not quantified in this paper.
  • domain assumption The Krämer modified Jacobian J*, which zeroes off-diagonal blocks, preserves accuracy (Section 2.3).
    The paper asserts 'extensive evidence' but provides no quantitative comparison in this manuscript. This assumption is needed for the O(d) linear scaling claim.
  • domain assumption The RKDP solver output is treated as the true ODE solution for benchmarking.
    Section 5.8: 'We shall assume that the output of the RKDP solver differs negligibly from the true ODE solution.' This is standard practice for numerical benchmarks.

how reviews work

0 comments
Cite this review

Pith. "Pith review of rodeo: Probabilistic Methods of Parameter Inference for Ordinary Differential Equations." pith.science (2026). https://pith.science/paper/EX3SOIIJ

@misc{pith2026250621776,
  author       = {Pith},
  title        = {Pith review of: rodeo: Probabilistic Methods of Parameter Inference for Ordinary Differential Equations},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/EX3SOIIJ}},
  note         = {Machine review of arXiv:2506.21776}
}
read the original abstract

Parameter estimation for ordinary differential equations (ODEs) plays a fundamental role in the analysis of dynamical systems. Generally lacking closed-form solutions, ODEs are traditionally approximated using deterministic solvers. However, there is a growing body of evidence to suggest that probabilistic ODE solvers produce more reliable parameter estimates by better accounting for numerical uncertainty. Here we present rodeo, a Python library providing a fast, lightweight, and extensible interface to a broad class of probabilistic ODE solvers, along with several associated methods for parameter inference. At its core, rodeo provides a probabilistic solver that scales linearly in both the number of evaluation points and system variables. Furthermore, by leveraging state-of-the-art automatic differentiation (AD) and just-in-time (JIT) compiling techniques, rodeo is shown across several examples to provide fast, accurate, and scalable parameter inference for a variety of ODE systems.

Figures

Figures reproduced from arXiv: 2506.21776 by the authors.

Figure 1
Figure 1. rodeo, Euler and the exact solution at N = 50, 100, 200 evaluation points for t ∈ [0, 10]. 100 draws and the analytical posterior mean were produced from rodeo. For this problem, a closed-form solution is given by x (0)(t) = 1 3 [PITH_FULL_IMAGE:figures/full_fig_p019_1.png] view at source ↗
Figure 2
Figure 2. Simulated ODE and noisy observations of the FN model. [PITH_FULL_IMAGE:figures/full_fig_p022_2.png] view at source ↗
Figure 3
Figure 3. Parameter posteriors for the FN model at different step sizes [PITH_FULL_IMAGE:figures/full_fig_p032_3.png] view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: The parameter posteriors of the Basic and DALTON approximations for the Hes1 [PITH_FULL_IMAGE:figures/full_fig_p036_4.png]
Figure 5
Figure 5. Figure 5: The parameter posteriors of the Basic and non-Gaussian DALTON methods for [PITH_FULL_IMAGE:figures/full_fig_p038_5.png]

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

62 extracted references · 50 canonical work pages

  1. [1]

    , " * write output.state after.block = add.period write newline

    ENTRY address archive author booktitle chapter collaboration doi edition editor eid eprint howpublished institution isbn issn journal key month note number numpages organization pages publisher school series title type url volume year label extra.label sort.label short.list INTEGERS output.state before.all mid.sentence after.sentence after.block FUNCTION ...

  2. [2]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 global.max substring 't := if while FUNCTION word.in bbl.in capitalize " " * FUNCT...

  3. [3]

    Numerical Solution of Ordinary Differential Equations

    Atkinson K, Han W, Stewart DE (2009). Numerical Solution of Ordinary Differential Equations. John Wiley & Sons. ISBN 978-0-470-04294-6

  4. [4]

    The D eep M ind JAX E cosystem

    Babuschkin I, Baumli K, Bell A, Bhupatiraju S, Bruce J, Buchlovsky P, Budden D, Cai T, Clark A, Danihelka I, Fantacci C, Godwin J, Jones C, Hemsley R, Hennigan T, Hessel M, Hou S, Kapturowski S, Keck T, Kemaev I, King M, Kunesch M, Martens L, Merzic H, Mikulik V, Norman T, Quan J, Papamakarios G, Ring R, Ruiz F, Sanchez A, Schneider R, Sezener E, Spencer ...

  5. [5]

    Gaussian processes for Bayesian estimation in ordinary Differential Equations

    Barber D, Wang Y (2014). Gaussian processes for Bayesian estimation in ordinary Differential Equations. In EP Xing, T Jebara (eds.), Proceedings of the 31st international conference on machine learning, volume 32 of Proceedings of machine learning research, pp. 1485--1493. PMLR, Bejing, China. ://proceedings.mlr.press/v32/barber14.html

  6. [6]

    A Comparison of Discrete Linear Filtering Algorithms

    Bierman GJ (1973). A Comparison of Discrete Linear Filtering Algorithms. IEEE Transactions on Aerospace and Electronic Systems, AES-9(1), 28--37. doi:10.1109/TAES.1973.309697

  7. [7]

    Efficient and Modular Implicit Differentiation

    Blondel M, Berthet Q, Cuturi M, Frostig R, Hoyer S, Llinares-L \'o pez F, Pedregosa F, Vert JP (2021). Efficient and Modular Implicit Differentiation. arXiv preprint arXiv:2105.15183

  8. [8]

    deBInfer: Bayesian inference for dynamical models of biological systems in R

    Boersch-Supan PH, Ryan SJ, Johnson LR (2017). deBInfer: Bayesian inference for dynamical models of biological systems in R. Methods in Ecology and Evolution, 8(4), 511--518. doi:https://doi.org/10.1111/2041-210X.12679. https://besjournals.onlinelibrary.wiley.com/doi/pdf/10.1111/2041-210X.12679 , ://besjournals.onlinelibrary.wiley.com/doi/abs/10.1111/2041-...

Show all 62 references
  1. [9]

    Calibrated Adaptive Probabilistic ODE Solvers

    Bosch N, Hennig P, Tronarp F (2021). Calibrated Adaptive Probabilistic ODE Solvers. In A Banerjee, K Fukumizu (eds.), Proceedings of The 24th International Conference on Artificial Intelligence and Statistics, volume 130 of Proceedings of Machine Learning Research, pp. 3466--3...

  2. [10]

    Pick-and-Mix Information Operators for Probabilistic ODE Solvers

    Bosch N, Tronarp F, Hennig P (2022). Pick-and-Mix Information Operators for Probabilistic ODE Solvers. In G Camps-Valls, FJR Ruiz, I Valera (eds.), Proceedings of The 25th International Conference on Artificial Intelligence and Statistics, volume 151 of Proceedings of Machine ...

  3. [11]

    JAX : composable transformations of P ython+ N um P y programs

    Bradbury J, Frostig R, Hawkins P, Johnson MJ, Leary C, Maclaurin D, Necula G, Paszke A, Vander P las J, Wanderman- M ilne S, Zhang Q (2018). JAX : composable transformations of P ython+ N um P y programs. ://github.com/google/jax

  4. [12]

    Numerical Methods for Ordinary Differential Equations

    Butcher JC (2008). Numerical Methods for Ordinary Differential Equations. John Wiley & Sons

  5. [13]

    Accelerating Bayesian inference over nonlinear differential equations with Gaussian processes

    Calderhead B, Girolami M, Lawrence ND (2009). Accelerating Bayesian inference over nonlinear differential equations with Gaussian processes. In Advances in neural information processing systems, pp. 217--224

  6. [14]

    Penalized Nonlinear Least Squares Estimation of Time-Varying Parameters in Ordinary Differential Equations

    Cao J, Huang JZ, Wu H (2012). Penalized Nonlinear Least Squares Estimation of Time-Varying Parameters in Ordinary Differential Equations. Journal of Computational and Graphical Statistics, 21(1), 42--56. ISSN 10618600. ://www.jstor.org/stable/23248822

  7. [15]

    Neural Ordinary Differential Equations

    Chen TQ, Rubanova Y, Bettencourt J, Duvenaud DK (2018). Neural Ordinary Differential Equations. In Advances in Neural Information Processing Systems , pp. 6571--6583

  8. [16]

    Bayesian solution uncertainty quantification for differential equations

    Chkrebtii OA, Campbell DA, Calderhead B, Girolami MA (2016). Bayesian solution uncertainty quantification for differential equations. Bayesian Analysis, 11(4), 1239--1267. ISSN 1936-0975. doi:10.1214/16-BA1017. ://projecteuclid.org/euclid.ba/1473276259

  9. [17]

    Statistical analysis of differential equations: introducing probability measures on numerical solutions

    Conrad PR, Girolami M, Särkkä S, Stuart A, Zygalakis K (2017). Statistical analysis of differential equations: introducing probability measures on numerical solutions. Statistics and Computing, 27(4), 1065--1082. ISSN 0960-3174, 1573-1375. doi:10.1007/s11222-016-9671-0. ://lin...

  10. [18]

    simode: Statistical Inference for Systems of Ordinary Differential Equations using Separable Integral-Matching

    Dattner I, Yaari R (2020). simode: Statistical Inference for Systems of Ordinary Differential Equations using Separable Integral-Matching. R package version 1.2.0, ://CRAN.R-project.org/package=simode

  11. [19]

    Bayesian numerical analysis

    Diaconis P (1988). Bayesian numerical analysis. In J Berger, S Gupta (eds.), Statistical Decision Theory and Related Topics IV , volume 1, pp. 163--175. Springer-Verlag, New York

  12. [20]

    ODE parameter inference using adaptive gradient matching with Gaussian processes

    Dondelinger F, Husmeier D, Rogers S, Filippone M (2013). ODE parameter inference using adaptive gradient matching with Gaussian processes. In CM Carvalho, P Ravikumar (eds.), Proceedings of the sixteenth international conference on artificial intelligence and statistics, volum...

  13. [21]

    A family of embedded Runge - Kutta formulae

    Dormand JR, Prince PJ (1980). A family of embedded Runge - Kutta formulae. Journal of Computational and Applied Mathematics, 6(1), 19--26. Publisher: Elsevier

  14. [22]

    Hybrid Monte Carlo

    Duane S, Kennedy AD, Pendleton BJ, Roweth D (1987). Hybrid Monte Carlo . Physics Letters B, 195(2), 216--222. doi:10.1016/0370-2693(87)91197-X

  15. [23]

    Bayesian Data Analysis

    Gelman A, Carlin JB, Stern HS, Dunson DB, Vehtari A, Rubin DB (2013). Bayesian Data Analysis . 3rd edition. Chapman & Hall, New York, NY. ISBN 978-0-429-11307-9

  16. [24]

    Fast approximate Bayesian computation for estimating parameters in differential equations

    Ghosh S, Dasmahapatra S, Maharatna K (2017). Fast approximate Bayesian computation for estimating parameters in differential equations. Statistics and Computing, 27(1), 19--38. ISSN 0960-3174, 1573-1375. doi:10.1007/s11222-016-9643-4. ://link.springer.com/10.1007/s11222-016-9643-4

  17. [25]

    Mixed Variational Inference

    Gianniotis N (2019). Mixed Variational Inference. In 2019 International Joint Conference on Neural Networks ( IJCNN ) , pp. 1--8. IEEE, Budapest, Hungary. ISBN 978-1-72811-985-4. doi:10.1109/IJCNN.2019.8852348. ://ieeexplore.ieee.org/document/8852348/

  18. [26]

    Scalable variational inference for dynamical systems

    Gorbach NS, Bauer S, Buhmann JM (2017). Scalable variational inference for dynamical systems. In Proceedings of the 31st international conference on neural information processing systems, NIPS '17, pp. 4809--4818. Curran Associates Inc., Red Hook, NY, USA. ISBN 978-1-5108-6096-4

  19. [27]

    Numerical Methods for Ordinary Differential Equations: Initial Value Problems

    Griffiths DF, Higham DJ (2010). Numerical Methods for Ordinary Differential Equations: Initial Value Problems. Springer Science & Business Media

  20. [28]

    Array programming with NumPy

    Harris CR, Millman KJ, van der Walt SJ, Gommers R, Virtanen P, Cournapeau D, Wieser E, Taylor J, Berg S, Smith NJ, Kern R, Picus M, Hoyer S, van Kerkwijk MH, Brett M, Haldane A, del R \' i o JF, Wiebe M, Peterson P, G \' e rard-Marchant P, Sheppard K, Reddy T, Weckesser W, Abb...

  21. [29]

    Probabilistic numerics and uncertainty in computations

    Hennig P, Osborne MA, Girolami M (2015). Probabilistic numerics and uncertainty in computations. Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences, 471(2179), 20150142. ISSN 1364-5021, 1471-2946. doi:10.1098/rspa.2015.0142. ://royalsocietypubl...

  22. [30]

    ODEPACK , a systematized collection of ODE solvers

    Hindmarsh AC (1983). ODEPACK , a systematized collection of ODE solvers. IMACS Transactions on Scientific Computation, 1, 55--64. Publisher: North-Holland

  23. [31]

    CollocInfer: Collocation Inference in Differential Equation Models

    Hooker G, Ramsay JO, Xiao L (2016). CollocInfer: Collocation Inference in Differential Equation Models. Journal of Statistical Software, 75(2), 1–52. doi:10.18637/jss.v075.i02. ://www.jstatsoft.org/index.php/jss/article/view/v075i02

  24. [32]

    Discrete square root filtering: A survey of current techniques

    Kaminski P, Bryson A, Schmidt S (1971). Discrete square root filtering: A survey of current techniques. IEEE Transactions on Automatic Control, 16(6), 727--736. doi:10.1109/TAC.1971.1099816

  25. [33]

    Active uncertainty calibration in Bayesian ODE solvers

    Kersting H, Hennig P (2016). Active uncertainty calibration in Bayesian ODE solvers. In 32nd conference on uncertainty in artificial intelligence ( UAI 2016) , pp. 309--318

  26. [34]

    Differentiable Likelihoods for Fast Inversion of Likelihood-Free Dynamical Systems

    Kersting H, Kr \"a mer N, Schiegg M, Daniel C, Tiemann M, Hennig P (2020 a ). Differentiable Likelihoods for Fast Inversion of Likelihood-Free Dynamical Systems. In HD III, A Singh (eds.), Proceedings of the 37th International Conference on Machine Learning, volume 119 of Proc...

  27. [35]

    Convergence rates of Gaussian ODE filters

    Kersting H, Sullivan T, Hennig P (2020 b ). Convergence rates of Gaussian ODE filters. Statistics and Computing, 30, 1--26. doi:10.1007/s11222-020-09972-4

  28. [36]

    On Neural Differential Equations

    Kidger P (2021). On Neural Differential Equations. Ph.D. thesis, University of Oxford. ://arxiv.org/pdf/2202.02435.pdf

  29. [37]

    Probabilistic ODE solutions in millions of dimensions

    Kr \"a mer N, Bosch N, Schmidt J, Hennig P (2021). Probabilistic ODE solutions in millions of dimensions. In Proceedings of the 39th International Conference on Machine Learning. ://proceedings.mlr.press/v162/kramer22b/kramer22b.pdf

  30. [38]

    Stable Implementation of Probabilistic ODE Solvers

    Kr \"a mer N, Hennig P (2020). Stable Implementation of Probabilistic ODE Solvers. doi:10.48550/ARXIV.2012.10106. ://arxiv.org/abs/2012.10106

  31. [39]

    a mer PN (2024). Implementing probabilistic numerical solvers for differential equations. Ph.D. thesis, Universit \

    Kr \"a mer PN (2024). Implementing probabilistic numerical solvers for differential equations. Ph.D. thesis, Universit \"a t T \"u bingen

  32. [40]

    B lackjax: A sampling library for JAX

    Lao J, Louf R (2020). B lackjax: A sampling library for JAX . ://github.com/blackjax-devs/blackjax

  33. [41]

    Multiphase MCMC sampling for parameter inference in nonlinear ordinary differential equations

    Lazarus A, Husmeier D, Papamarkou T (2018). Multiphase MCMC sampling for parameter inference in nonlinear ordinary differential equations. In A Storkey, F Perez-Cruz (eds.), Proceedings of the twenty-first international conference on artificial intelligence and statistics, vol...

  34. [42]

    deGradInfer: Parameter Inference for Systems of Differential Equation

    Macdonald B, Dondelinger F (2020). deGradInfer: Parameter Inference for Systems of Differential Equation. R package version 1.0.1, ://CRAN.R-project.org/package=deGradInfer

  35. [43]

    A Practical Bayesian Framework for Backpropagation Networks

    MacKay DJC (1992). A Practical Bayesian Framework for Backpropagation Networks. Neural Computation, 4(3), 448--472. ISSN 0899-7667, 1530-888X. doi:10.1162/neco.1992.4.3.448. ://direct.mit.edu/neco/article/4/3/448-472/5654

  36. [44]

    Bayesian estimation of time-varying parameters in ordinary differential equation models with noisy time-varying covariates

    Meng L, Zhang J, Zhang X, and GF (2021). Bayesian estimation of time-varying parameters in ordinary differential equation models with noisy time-varying covariates. Communications in Statistics - Simulation and Computation, 50(3), 708--723. doi:10.1080/03610918.2019.1565584. h...

  37. [45]

    R package for statistical inference in dynamical systems using kernel based gradient matching: KGode

    Niu M, Wandy J, Daly R, Rogers S, Husmeier D (2021). R package for statistical inference in dynamical systems using kernel based gradient matching: KGode. Computational Statistics, 36(1), 715--747. doi:10.1007/s00180-020-01014-. ://ideas.repec.org/a/spr/compst/v36y2021i1d10.10...

  38. [46]

    Numerical Optimization

    Nocedal J, Wright SJ (2006). Numerical Optimization. 2e edition. Springer, New York, NY, USA

  39. [47]

    Automatic selection of methods for solving stiff and nonstiff systems of ordinary differential equations

    Petzold L (1983). Automatic selection of methods for solving stiff and nonstiff systems of ordinary differential equations. SIAM Journal on Scientific and Statistical Computing, 4(1), 136--148. Publisher: SIAM

  40. [48]

    Multi-Level Modeling of Early COVID -19 Epidemic Dynamics in French Regions and Estimation of the Lockdown Impact on Infection Rate

    Prague M, Wittkop L, Collin A, Dutartre D, Clairon Q, Moireau P, Thi \'e baut R, Hejblum BP (2020). Multi-Level Modeling of Early COVID -19 Epidemic Dynamics in French Regions and Estimation of the Lockdown Impact on Infection Rate. Preprint, Epidemiology . doi:10.1101/2020.04...

  41. [49]

    Probabilistic ODE solvers with Runge - Kutta means

    Schober M, Duvenaud DK, Hennig P (2014). Probabilistic ODE solvers with Runge - Kutta means. In Advances in neural information processing systems, pp. 739--747

  42. [50]

    A probabilistic model for the numerical solution of initial value problems

    Schober M, Särkkä S, Hennig P (2019). A probabilistic model for the numerical solution of initial value problems. Statistics and Computing, 29(1), 99--122. ISSN 0960-3174, 1573-1375. doi:10.1007/s11222-017-9798-7. ://link.springer.com/10.1007/s11222-017-9798-7

  43. [51]

    FitzHugh--Nagumo Model , pp

    Sherwood WE (2013). FitzHugh--Nagumo Model , pp. 1--11. Springer New York, New York, NY. ISBN 978-1-4614-7320-6. doi:10.1007/978-1-4614-7320-6_147-1. ://doi.org/10.1007/978-1-4614-7320-6_147-1

  44. [52]

    Bayesian solution of ordinary differential equations

    Skilling J (1992). Bayesian solution of ordinary differential equations. In CR Smith, GJ Erickson, PO Neudorfer (eds.), Maximum Entropy and Bayesian Methods : Seattle , 1991 , Fundamental Theories of Physics , pp. 23--37. Springer Netherlands, Dordrecht. ISBN 978-94-017-2219-3...

  45. [53]

    Stan Modeling Language Users Guide and Reference Manual

    Stan Development Team (2023). Stan Modeling Language Users Guide and Reference Manual. ://mc-stan.org

  46. [54]

    Probabilistic linear multistep methods

    Teymur O, Zygalakis K, Calderhead B (2016). Probabilistic linear multistep methods. In Advances in Neural Information Processing Systems 29, pp. 4321--4328. Curran Associates, Inc

  47. [55]

    Fenrir : Physics-Enhanced Regression for Initial Value Problems

    Tronarp F, Bosch N, Hennig P (2022). Fenrir : Physics-Enhanced Regression for Initial Value Problems. In K Chaudhuri, S Jegelka, L Song, C Szepesvari, G Niu, S Sabato (eds.), Proceedings of the 39th International Conference on Machine Learning, volume 162 of Proceedings of Mac...

  48. [56]

    Probabilistic solutions to ordinary differential equations as non-linear Bayesian filtering: a new perspective

    Tronarp F, Kersting H, Särkkä S, Hennig P (2018). Probabilistic solutions to ordinary differential equations as non-linear Bayesian filtering: a new perspective. arXiv:1810.03440 [stat]. ArXiv: 1810.03440, ://arxiv.org/abs/1810.03440

  49. [57]

    Bayesian ODE solvers: the maximum a posteriori estimate

    Tronarp F, S \"a rkk \"a S, Hennig P (2021). Bayesian ODE solvers: the maximum a posteriori estimate. Statistics and Computing, 31(3), 23. ISSN 1573-1375. doi:10.1007/s11222-021-09993-7. ://doi.org/10.1007/s11222-021-09993-7

  50. [58]

    pCODE: Estimation of an Ordinary Differential Equation Model by Parameter Cascade Method

    Wang H, Cao J (2022). pCODE: Estimation of an Ordinary Differential Equation Model by Parameter Cascade Method. R package version 0.9.4, ://CRAN.R-project.org/package=pCODE

  51. [59]

    ProbNum: Probabilistic Numerics in Python

    Wenger J, Krämer N, Pförtner M, Schmidt J, Bosch N, Effenberger N, Zenn J, Gessner A, Karvonen T, Briol FX, Mahsereci M, Hennig P (2021). ProbNum: Probabilistic Numerics in Python . ://arxiv.org/abs/2112.02100

  52. [60]

    Fast Gaussian process based gradient matching for parameter identification in systems of nonlinear ODEs

    Wenk P, Gotovos A, Bauer S, Gorbach NS, Krause A, Buhmann JM (2019). Fast Gaussian process based gradient matching for parameter identification in systems of nonlinear ODEs . In K Chaudhuri, M Sugiyama (eds.), Proceedings of the twenty-second international conference on artifi...

  53. [61]

    Data-Adaptive Probabilistic Likelihood Approximation for Ordinary Differential Equations

    Wu M, Lysy M (2024). Data-Adaptive Probabilistic Likelihood Approximation for Ordinary Differential Equations. In S Dasgupta, S Mandt, Y Li (eds.), Proceedings of The 27th International Conference on Artificial Intelligence and Statistics, volume 238 of Proceedings of Machine ...

  54. [62]

    Inference of dynamic systems from noisy and sparse data via manifold-constrained Gaussian processes

    Yang S, Wong SWK, Kou SC (2021). Inference of dynamic systems from noisy and sparse data via manifold-constrained Gaussian processes. Proceedings of the National Academy of Sciences, 118(15), e2020397118. ISSN 0027-8424, 1091-6490. doi:10.1073/pnas.2020397118. ://www.pnas.org/...

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.