betaFeedback

WorksMathematics

Entire sig-jump-diffusions, their expected signature and applications to score-based generative modeling

Abstract

We develop score-based generative models for time series and standard finite-dimensional data, building on the analytical computation of the expected signature of a new class of jump-diffusions, called entire sig-jump-diffusions. These are semimartingales whose characteristics are given by linear combinations of potentially infinitely many components of their own signature. We show that their expected signature can be computed by solving linear ODEs in the extended tensor algebra, providing a tractable foundation for learning procedures. Specifically, the learning of a generalized score function for jump diffusions, a key ingredient of our generative modeling framework, can be reduced to solving linear ODEs and a deterministic optimization problem. We further prove that polynomial jump-diffusions can be embedded into this framework, and that their expected signature reduces to a finite-dimensional system of linear ODEs. Building on this result, we develop an alternative approach for finite-dimensional data, based on inverse signature matching, a method that learns the score by matching the expected signature of the reverse process with that of the inverse forward signature.

MSC (2020) Classification: 60L10, 60J76, 68T05

Introduction

Score-based generative models have become a leading paradigm for sampling from complex distributions Sohl-Dickstein et al. [65]; Song and Ermon [66]; Ho et al. (2020); Song et al. [67]; Tang and Zhao [69]. Extending them to time series data is challenging, as sequential data are path-valued. Indeed, applying standard score-based generative modeling via stochastic differential equations to time series generation requires considering time-reversed (jump) diffusions on path space, which are inherently infinite-dimensional and potentially intractable. A natural way to overcome this complexity is to handle paths through their signature, a universal feature map given by iterated integrals Lyons [54]; Chevyrev and Kormilitzin [16]; Cass and Salvi [14], which – in its logarithmic form – has recently been applied by Barancikova et al. [5] in the context of generative time series modeling. In this paper, we construct score-based generative models for time series as well as finite-dimensional data, building on an explicit representation of the expected signature of a broad class of jump-diffusions. Our main contributions are the following.

  • Entire sig-jump-diffusions. We introduce the class of entire sig-jump-diffusions, whose expected signature admits an explicit representation via linear ordinary differential equations (Definition 3.3, Theorem 3.13).

  • Score-based generative modeling with (lifted) entire sig-jump-diffusions for time series. We exploit this representation to construct generative models for time series, in which a generalized score function is learned through a deterministic optimization problem (Sections 4.1–4.4, Theorem 4.9).

  • Finite-dimensional data generation and inverse signature matching. For data in Rd\mathbb{R}^{d}, we adapt this framework to the subclass of polynomial processes introduced in Cuchiero et al. [22]; Filipović and Larsson [33] (Remark 4.10). Additionally we propose an alternative to score matching, called inverse signature matching (§ 4.5).

Our starting point is the expected signature, the path-space analogue of the sequence of moments of a random variable, which under suitable conditions characterizes the law of a stochastic process, Chevyrev and Lyons [17]; Chevyrev and Oberhauser [18]. It is therefore a natural tool whenever laws on path space must be estimated or compared, as in parameter estimation Papavasiliou and Ladroue [61]; Semnani et al. [64], hypothesis testing Gasteratos et al. [36] and generative modeling of time series Liao et al. [52]; Lou et al. [53]. It often also reduces stochastic control problems Kalsi et al. [48]; Bayer et al. [9]; Cuchiero and Möller [23]; Gennaro et al. [38]; Abi Jaber et al. [1] and financial derivative pricing Lyons et al. [55]; Cuchiero et al. [21]; Abi Jaber et al. [2] to simple deterministic solutions, since expected costs and prices become linear functionals of the expected signature.

For these reasons, the computation of expected signatures has received considerable attention. Explicit or recursive representations are available for Brownian motion Fawcett [31]; Lyons and Victoir [57]; Lyons and Ni [56]; Boedihardjo et al. [10], Gaussian processes Boedihardjo et al. [11]; Cass and Ferrucci [13], Lévy processes Friz and Shekhar [35]; Cuchiero et al. [25], and more general diffusion and semimartingale models Ni [60]; Friz et al. [34]; Cuchiero et al. [27].

Entire sig-jump-diffusions. We contribute to this line of work by introducing a general class of jump-diffusions whose expected signature is obtained by solving a linear ODE in the extended tensor algebra. To this end, we define entire sig-jump-diffusions as semimartingales whose characteristics can be written as a linear combination of potentially infinitely many components of their own signature (Definition 3.3). This framework is very general: by the universal approximation properties of the signature, continuous path functionals can be approximated arbitrarily well, on suitable compact sets and with respect to suitable variation metrics, by linear functionals of the signature [48, 9, 25], therefore approximating (in suitable metrics) any semimartingale with sufficiently regular characteristics. The class extends signature SDEs introduced in Cuchiero et al. [27], where only semimartingales with continuous trajectories are considered. In particular, it includes the signature models for asset pricing of [62, 21, 20 b].

Based on duality arguments for stochastic processes [47, 30], as in [27 a], we show that expectations of potentially infinitely many linear combinations of the signature components of such processes can be computed by solving linear ODEs in the extended tensor algebra (Theorem 3.13), extending to the non-continuous setting the results of Cuchiero et al. [27]. The main technical challenge is the convergence of the resulting infinite series, which is considerably harder in the presence of jumps, as it requires control of the series and all its derivatives (equation (5)).

For certain important cases, we can prove that these convergence conditions are satisfied. This applies to holomorphic jump-diffusions as of Cuchiero et al. [26] (Subsection 3.3.1), to polynomial processes introduced in [22, 33] (Subsection 3.3.3), and to some jump-diffusions with possibly path-dependent characteristics (Remark 3.27), all of which embed in the entire sig-jump-diffusion framework. In particular, the truncated signature of a polynomial jump-diffusion is again a finite-dimensional polynomial jump-diffusion (Proposition 3.24), so that its expected signature reduces to an easily computable matrix exponential (Example 3.28). Another tractable subclass, whose truncated signature lift is a finite-dimensional Markov process, is that of entire sig-jump-diffusions with truncated signature-dependent coefficients (Subsection 3.3.2).

Score-based generative modeling with (lifted) entire sig-jump-diffusions for time series. We apply the framework of entire sig-jump-diffusions, specifically the subclass of Subsection 3.3.2, to score-based generative modeling of time series. We encode the distribution of the observed time series through the distribution of their truncated signatures and perturb it with a lifted entire sig-jump-diffusion converging to an invariant measure independent of the initial data distribution. Starting from this invariant distribution, a suitable reverse-time process generates new samples approximating the original signature distribution.

The key ingredient in this construction is a generalized score function, which extends the classical score function to jump-diffusions. In the Lévy setting it corresponds to the score considered in Baule [8] for finite jump activity and in Yoon et al. [71] for isotropic α\alpha-stable noise. The latter has been taken up by Popov et al. [63], who use a parametrized family of reverse-time processes sharing the prescribed marginal distributions (and not the whole law on path space), a principle that we extend beyond the α\alpha-stable case to general jump-diffusions. The crucial difference to these previous works is that our definition of the generalized score function is formulated directly in terms of the generator and the marginal probability measures, without requiring a particular structure of the jump mechanism. As shown in Sections 4.1 and § 4.2, this makes it applicable to state-dependent jump-diffusions and, in particular, to the lifted truncated signature dynamics of Subsection 3.3.2.

Our main contribution is to exploit the analytic structure of entire sig-jump-diffusions and their explicit expected signature representation to reduce the learning of this score to linear ODEs and a deterministic optimization problem (Section 4.3, see in particular Theorem 4.9). As already mentioned, our approach is related to the recent work of Barancikova et al. [5], who introduced score-based generative models operating on log-signature representations of time series. While their approach relies on diffusion processes in the vector space of log-signatures, we consider lifted entire sig-jump-diffusions evolving directly on the space of signatures, which provides a natural connection to the stochastic dynamics of the underlying time series and allows the coefficients of the forward process to be calibrated to the observed dynamics. We refer to § 4 for a more detailed explanation. Beyond this, our approach differs from Barancikova et al. [5] in two further respects. First, we incorporate jumps into the generative dynamics. Second, thanks to the analytic tractability of entire sig-jump-diffusions, the score is learned through the explicit expected signature instead of neural network training.

To show that suitable forward processes exist, in Section 4.4 we construct an entire sig-diffusion whose truncated signature lift converges to a unique invariant measure, independently of the initial data distribution, and thus provides a natural forward perturbation process.

Since the framework operates on signatures rather than on time series, recovering a time series from a generated signature requires a separate reconstruction step, for which several inversion procedures exist Lyons and Xu [58, 59]; Fermanian et al. [32]; Barancikova et al. [5]. As truncated signatures do not determine paths uniquely, this step is subject to approximation issues, which we do not address here.

Finite-dimensional data generation and inverse signature matching. The entire sig-jump-diffusion framework can also be used to generate finite-dimensional data, i.e., new samples of random variables with values in Rd\mathbb{R}^{d}. To this end, we use polynomial processes Cuchiero et al. [22]; Filipović and Larsson [33], analyzed in Subsection 3.3.3, for the forward dynamics. In this setting, learning the generalized score function reduces to computing moments and solving, again, a deterministic optimization problem (Remark 4.10).

We also propose an alternative to standard score matching, which we call inverse signature matching, tailored to polynomial diffusions. Since, under suitable conditions, the expected signature characterizes the law of a stochastic process Chevyrev and Oberhauser [18], the key idea is to learn the generalized score by matching the expected signature of the reverse process to that of the inverse forward signature. Under a suitable parametrization of the score map, the polynomial structure ensures that both expectations can be computed analytically, which reduces the problem again to a deterministic optimization problem (§ 4.5).

Finally, we stress that the goal of the paper is to establish the framework and the analytic learning procedure, while a numerical investigation of the generative models, on time series as well as on finite-dimensional data, is left to future work.

Related work on generative modeling. As the literature on expected signatures was already discussed above, we here only focus on generative modeling. In score-based generative models, the reverse-time drift involves the score of the marginal densities Haussmann and Pardoux [43], which is usually learned by (denoising) score matching Hyvärinen [45]; Vincent [70]. Besides the reverse SDE, the probability-flow ODE Song et al. [67] gives a deterministic sampler with the same marginal laws. Jump noise has been considered in Yoon et al. [71]; Baule [8], with a family of marginal-preserving reverse dynamics in Popov et al. [63]; we compare with these works in the description of our contributions above. Other works build on pretrained diffusion models, for instance for conditional generation and fine-tuning Han et al. [42]; Guo et al. [40] or distillation Ge et al. (2026); these are complementary to the construction of the forward and reverse dynamics studied here.

For time series, several generative models are based on signatures Liao et al. [52]; Lou et al. [53], the closest to our work being Barancikova et al. [5], which we already discussed in detail above. Other approaches that do not rely on signatures include the Schrödinger bridge model of Hamdouche et al. [41], whose path-dependent drift is estimated nonparametrically, its extension to jump-diffusions De Marco et al. [28], and the sequential forward-backward diffusion of Cao et al. [12], which generates adapted series by conditioning on the generated history and comes with statistical guarantees.

Organization of the paper. In Section 2, we introduce the algebraic setting of signature theory and the notion of signature for càdlàg semimartingales. In Section 3, we introduce the class of entire sig-jump-diffusions, present the main representation results (Theorem 3.13), and show that holomorphic jump-diffusions, entire sig-jump-diffusions with truncated signature-dependent coefficients, and polynomial processes can be embedded within this framework. § 4 is dedicated to the application to score-based generative modeling.

Preliminaries

Notation

Fix d∈Nd \in\mathbb{N} and let Rd\mathbb{R}^{d} be the Euclidean space. The extended tensor algebra over Rd\mathbb{R}^{d} is defined by

T(Rd):=∏n=0∞(Rd)⊗n,T\left(\mathbb{R}^{d}\right) := \prod_{n=0}^{\infty}\left(\mathbb{R}^{d}\right)^{\otimes n},

where (Rd)⊗n\left(\mathbb{R}^{d}\right)^{\otimes n} denotes the nn-fold tensor product of Rd\mathbb{R}^{d} with the convention (Rd)⊗0:=R\left(\mathbb{R}^{d}\right)^{\otimes0} := \mathbb{R}. We equip T((Rd))T\left(\left(\mathbb{R}^{d}\right)\right) with the standard addition ++, tensor multiplication ⊗\otimes and scalar multiplication. For N∈NN \in\mathbb{N}, the truncated tensor algebra is defined by

TN(Rd):=⨁n=0N(Rd)⊗n.T^{N}\left(\mathbb{R}^{d}\right) := \bigoplus_{n=0}^{N}\left(\mathbb{R}^{d}\right)^{\otimes n}.

Let I=(i1,…,in)I=(i_{1},\ldots,i_{n}) be a multi-index with entries in {1,…,d}\{1,\ldots,d\}. Denoting by ϵ1,…,ϵd\epsilon_{1},\ldots,\epsilon_{d} the canonical basis of Rd\mathbb{R}^{d}, we use the notation ∣I∣:=n|I|:=n and ϵI:=ϵi1⊗ϵi2⊗⋯⊗ϵin\epsilon_{I}:=\epsilon_{i_{1}}\otimes\epsilon_{i_{2}}\otimes\cdots\otimes\epsilon_{i_{n}}. Observe that (ϵI)I(\epsilon_{I})_{I} is the canonical orthonormal basis of (Rd)⊗n(\mathbb{R}^{d})^{\otimes n}. Furthermore, we denote by ϵ∅\epsilon_{\varnothing} the basis element of (Rd)⊗0(\mathbb{R}^{d})^{\otimes0} and set ∣∅∣:=0|\varnothing|:=0. We also set I′:=(i1,…,in−1)I^{\prime}:=(i_{1},\ldots,i_{n-1}) for n>1n>1, I′=∅I^{\prime}=\varnothing for n=1n=1, I′′:=(I′)′I^{\prime\prime}:=(I^{\prime})^{\prime} for n>1n>1, and use the convention ϵI′=0\epsilon_{I^{\prime}}=0 for I=∅I=\varnothing. For u∈T((Rd))\mathbf{u}\in T((\mathbb{R}^{d})), we denote its projection on TN(Rd)T^{N}(\mathbb{R}^{d}) by u≤N\mathbf{u}^{\leq N} and let uIu_{I} denote its component corresponding to ϵI\epsilon_{I}.

For two multi-indices I∈{1,…,d}∣I∣I\in\{1,\ldots,d\}^{|I|}, J∈{1,…,d}∣J∣J\in\{1,\ldots,d\}^{|J|}, and two indices of length 1, a,b∈{1,…,d}a,b\in\{1,\ldots,d\}, the shuffle product ⊔\sqcup is defined recursively by

ϵI⊔ϵ∅=ϵ∅⊔ϵI=ϵI,\epsilon_{I}\sqcup\epsilon_{\varnothing}=\epsilon_{\varnothing}\sqcup\epsilon_{I}=\epsilon_{I},
ϵIa⊔ϵJb=(ϵI⊔ϵJb)⊗ϵa+(ϵIa⊔ϵJ)⊗ϵb,\epsilon_{Ia}\sqcup\epsilon_{Jb}=(\epsilon_{I}\sqcup\epsilon_{Jb})\otimes\epsilon_{a}+(\epsilon_{Ia}\sqcup\epsilon_{J})\otimes\epsilon_{b},

where IaIa denotes the concatenation of multi-indices.

For u∈T(Rd)\mathbf{u}\in T(\mathbb{R}^{d}) and x∈T((Rd))\mathbf{x}\in T((\mathbb{R}^{d})), we write

⟨u,x⟩:=∑n=0∞∑∣I∣=nuIxI.\langle\mathbf{u},\mathbf{x}\rangle:=\sum_{n=0}^{\infty}\sum_{|I|=n}u_{I}x_{I}.

Define GN(Rd):={u≤N∈TN(Rd):u∈G((Rd))}G^{N}(\mathbb{R}^{d}):=\{\mathbf{u}^{\leq N}\in T^{N}(\mathbb{R}^{d}):\mathbf{u}\in G((\mathbb{R}^{d}))\} for

G((Rd)):={x∈T((Rd)):⟨ϵ∅,x⟩=1, ⟨ϵI⊔ϵJ,x⟩=⟨ϵI,x⟩⟨ϵJ,x⟩, for all I,J},G((\mathbb{R}^{d})):=\{\mathbf{x}\in T((\mathbb{R}^{d})):\langle\epsilon_{\varnothing},\mathbf{x}\rangle=1,\ \langle\epsilon_{I}\sqcup\epsilon_{J},\mathbf{x}\rangle=\langle\epsilon_{I},\mathbf{x}\rangle\langle\epsilon_{J},\mathbf{x}\rangle,\ \text{for all }I,J\},

and for S⊆RdS\subseteq\mathbb{R}^{d}, set

S:={x∈G((Rd)):∑∣I∣=1xIϵI∈S}.(1)S:=\left\{\mathbf{x}\in G((\mathbb{R}^{d})):\sum_{|I|=1}x_{I}\epsilon_{I}\in S\right\}. \tag*{(1)}

The set SS is endowed with the sigma-algebra induced by the product sigma-algebra on T((Rd))T((\mathbb{R}^{d})).

Define

S∗:={u∈T((Rd))+iT((Rd)):∣u∣x<∞ for all x∈S},S^{*}:=\{\mathbf{u}\in T((\mathbb{R}^{d}))+iT((\mathbb{R}^{d})):\lvert\mathbf{u}\rvert_{\mathbf{x}}<\infty\text{ for all }\mathbf{x}\in S\},

where ∣u∣x:=∑n=0∞∑∣I∣=n∣uIxI∣\lvert\mathbf{u}\rvert_{\mathbf{x}}:=\sum_{n=0}^{\infty}\sum_{\lvert I\rvert=n}\lvert u_{I}x_{I}\rvert. For u∈S∗\mathbf{u}\in S^{*}, we again write

⟨u,x⟩:=∑n=0∞∑∣I∣=nuIxI.\langle\mathbf{u},\mathbf{x}\rangle:=\sum_{n=0}^{\infty}\sum_{\lvert I\rvert=n}u_{I}x_{I}.

Moreover, we set u(0):=u\mathbf{u}^{(0)}:=\mathbf{u} and for k∈Nk\in\mathbb{N}, ∣I∣=k\lvert I\rvert=k, we define uI(k)∈T((Rd))+iT((Rd))\mathbf{u}_{I}^{(k)}\in T((\mathbb{R}^{d}))+iT((\mathbb{R}^{d})) via

uI(k):=∑∣J∣≥0uJI.\mathbf{u}_{I}^{(k)}:=\sum_{\lvert J\rvert\geq0}u_{JI}.

References

  1. [1]Abi Jaber, E., Attal, E., and Sotnikov, D. (2026). Stochastic control with signatures via Riccati equations on the tensor algebra. Preprint, arXiv:2607.03986.arxiv.org/abs/2607.03986
  2. [2]Abi Jaber, E., Hainaut, D., and Motte, E. (2025). Signature Approach for Pricing and Hedging Path-Dependent Options with Frictions. Preprint, arXiv:2511.23295.arxiv.org/abs/2511.23295
  3. [3]Abi Jaber, E. and Sotnikov, D. (2025). Exponentially Fading Memory Signature. Preprint, arXiv:2507.03700.arxiv.org/abs/2507.03700
  4. [4]Arnaudon, M. and Thalmaier, A. (2010). The Differentiation of Hypoelliptic Diffusion Semigroups. Illinois Journal of Mathematics, 54(4):1285–1311.arxiv.org/abs/1004.2174
  5. [5]Barancikova, B., Huang, Z., and Salvi, C. (2025). SigDiffusions: Score-Based Diffusion Models for Time Series via Log-Signature Embeddings. In The Thirteenth International Conference on Learning Representations.arxiv.org/abs/2406.10354
  6. [6]Barilari, D., Boscain, U., and Neel, R. W. (2019). Heat Kernel Asymptotics on Sub-Riemannian Manifolds with Symmetries and Applications to the Bi-Heisenberg Group. Annales de la Faculté des Sciences de Toulouse, 28(4):707–732.arxiv.org/abs/1606.01159
  7. [7]Baudoin, F., Hairer, M., and Teichmann, J. (2008). Ornstein–uhlenbeck processes on lie groups. Journal of Functional Analysis, 255(4):877–890.arxiv.org/abs/0711.2419
  8. [8]Baule, A. (2026). Generative modelling with jump-diffusions. Journal of Statistical Mechanics: Theory and Experiment, 2026(5):054002.
  9. [9]Bayer, C., Hager, P. P., Riedel, S., and Schoenmakers, J. (2023). Optimal stopping with signatures. The Annals of Applied Probability, 33(1):238–273.arxiv.org/abs/2105.00778
  10. [10]Boedihardjo, H., Diehl, J., Mezzarobba, M., and Ni, H. (2021). The expected signature of Brownian motion stopped on the boundary of a circle has finite radius of convergence. Bulletin of the London Mathematical Society, 53(1):285–299.DOI
  11. [11]Boedihardjo, H., Papavasiliou, A., and Qian, Z. (2013). Expected Signature of Gaussian Processes with Strictly Regular Kernels. Preprint, arXiv:1304.4930.arxiv.org/abs/1304.4930
  12. [12]Cao, H., Chen, M., Han, Y., and Xu, R. (2026). Diffusion Models for Adaptive Sequential Data Generation. Preprint, arXiv:2606.06007.arxiv.org/abs/2606.06007
  13. [13]Cass, T. and Ferrucci, E. (2024). On the Wiener Chaos Expansion of the Signature of a Gaussian Process. Probability Theory and Related Fields, 189:909–947.arxiv.org/abs/2207.08422
  14. [14]Cass, T. and Salvi, C. (2024). Lecture notes on rough paths and applications to machine learning. Preprint, arXiv:2404.06583.arxiv.org/abs/2404.06583
  15. [15]Chevyrev, I. and Friz, P. K. (2019). Canonical RDEs and general semimartingales as rough paths. The Annals of Probability, 47(1):420–463.DOI
  16. [16]Chevyrev, I. and Kormilitzin, A. (2026). A primer on the signature method in machine learning. In Signature Methods in Finance: An Introduction with Computational Applications, pages 3–64. Springer.
  17. [17]Chevyrev, I. and Lyons, T. (2016). Characteristic Functions of Measures on Geometric Rough Paths. The Annals of Probability, 44(6):4049–4082.DOI
  18. [18]Chevyrev, I. and Oberhauser, H. (2022). Signature Moments to Characterize Laws of Stochastic Processes. Journal of Machine Learning Research, 23(176):1–42.
  19. [19]Conforti, G. and Léonard, C. (2022). Time reversal of Markov processes with jumps under a finite entropy condition. Stochastic Processes and their Applications, 144:85–124.arxiv.org/abs/2104.10948
  20. [20]Cuchiero, C., Gazzani, G., Möller, J., and Svaluto-Ferro, S. (2025a). Joint calibration to SPX and VIX options with signature-based models. Mathematical Finance, 35(1):161–213.arxiv.org/abs/2301.13235
  21. [21]Cuchiero, C., Gazzani, G., and Svaluto-Ferro, S. (2023). Signature-Based Models: Theory and Calibration. SIAM Journal on Financial Mathematics, 14(3):910–957.arxiv.org/abs/2207.13136
  22. [22]Cuchiero, C., Keller-Ressel, M., and Teichmann, J. (2012). Polynomial processes and their applications to mathematical finance. Finance and Stochastics, 16(4):711–740.
  23. [23]Cuchiero, C. and Möller, J. (2025). Signature methods in stochastic portfolio theory. SIAM Journal on Financial Mathematics, 16(4):1239–1303.arxiv.org/abs/2310.02322
  24. [25]Cuchiero, C., Primavera, F., and Svaluto-Ferro, S. (2025b). Universal approximation theorems for continuous functions of càdlàg paths and Lévy-type signature models. Finance and Stochastics, 29(2):289–342.arxiv.org/abs/2208.02293
  25. [26]Cuchiero, C., Primavera, F., and Svaluto-Ferro, S. (2026a). Holomorphic jump-diffusions. Stochastic Processes and their Applications, 191:104781.arxiv.org/abs/2411.04238
  26. [27]Cuchiero, C., Svaluto-Ferro, S., and Teichmann, J. (2026b). Signature SDEs from an affine and polynomial perspective. Electronic Journal of Probability, 31:1–57.DOI
  27. [28]De Marco, S., Pham, H., and Zanni, D. (2026). Schrödinger bridges with jumps for time series generation. Preprint, arXiv:2602.20011.arxiv.org/abs/2602.20011
  28. [29]Doob, J. L. (1948). Asymptotic Properties of Markoff Transition Probabilities. Transactions of the American Mathematical Society, 63(3):393–421.
  29. [30]Ethier, S. N. and Kurtz, T. G. (1986). Markov processes: characterization and convergence. Wiley Series in Probability and Mathematical Statistics: Probability and Mathematical Statistics. John Wiley & Sons Inc.
  30. [31]Fawcett, T. (2003). Problems in Stochastic Analysis: Connections between Rough Paths and Non-Commutative Harmonic Analysis. PhD thesis, University of Oxford.DOI
  31. [32]Fermanian, A., Chang, J., Lyons, T., and Biau, G. (2024). The insertion method to invert the signature of a path. In Recent Advances in Econometrics and Statistics: Festschrift in Honour of Marc Hallin, pages 575–595. Springer.arxiv.org/abs/2304.01862
  32. [33]Filipović, D. and Larsson, M. (2020). Polynomial Jump-Diffusion Models. Stochastic Systems, 10(1):71–97.
  33. [34]Friz, P. K., Hager, P. P., and Tapia, N. (2026). On Expected Signatures and Signature Cumulants in Semimartingale Models. In Signature Methods in Finance: An Introduction with Computational Applications, pages 381–424. Springer.
  34. [35]Friz, P. K. and Shekhar, A. (2017). General Rough Integration, Lévy Rough Paths and a Lévy–Kintchine-Type Formula. The Annals of Probability, 45(4):2707–2765.arxiv.org/abs/1212.5888
  35. [36]Gasteratos, I., Jacquier, A., L ermercier, M., Lyons, T., and Salvi, C. (2025). Novelty detection on path space. Preprint, arXiv:2512.03243.arxiv.org/abs/2512.03243
  36. [38]Gennaro, A., M astrolia, T., and Primavera, F. (2026). Signature Methods for Optimal Market Making. Preprint, arXiv:2606.19772.arxiv.org/abs/2606.19772
  37. [39]Gerlach, M. and Nittka, R. (2012). A New Proof of Doob’s Theorem. Journal of Mathematical Analysis and Applications, 388(2):763–774.DOI
  38. [40]Guo, Z., Tang, W., and Xu, R. (2026). Conditional Diffusion Guidance under Hard Constraint: A Stochastic Analysis Approach. Preprint, arXiv:2602.05533.arxiv.org/abs/2602.05533
  39. [41]Hamdouche, M., Henry-Labordère, P., and Pham, H. (2026). Nonparametric generative modeling for time series via Schrödinger bridge. Journal of Machine Learning Research, 27(58):1–23.
  40. [42]Han, Y., Razaviyayn, M., and Xu, R. (2025). Stochastic control for fine-tuning diffusion models: Optimality, regularity, and convergence. In Proceedings of the 42nd International Conference on Machine Learning, volume 267 of Proceedings of Machine Learning Research, pages 21844–21870.arxiv.org/abs/2412.18164
  41. [43]Haussmann, U. G. and Pardoux, E. (1986). Time reversal of diffusions. The Annals of Probability, 14(4):1188–1205.DOI
  42. [45]Hyvärinen, A. (2005). Estimation of Non-Normalized Statistical Models by Score Matching. Journal of Machine Learning Research, 6:695–709.DOI
  43. [46]Jacod, J. and Shiryaev, A. (1987). Limit Theorems for Stochastic Processes. Springer Berlin, Heidelberg.
  44. [47]Jansen, S. and Kurt, N. (2014). On the notion(s) of duality for Markov processes. Probability Surveys, 11:59–120.arxiv.org/abs/1210.7193
  45. [48]Kalsi, J., Lyons, T., and Perez Arribas, I. (2020). Optimal execution with rough path signatures. SIAM Journal on Financial Mathematics, 11(2):470–493.arxiv.org/abs/1905.00728
  46. [49]Kurtz, T. G., Pardoux, E., and Protter, P. (1995). Stratonovich stochastic differential equations driven by general semimartingales. Annales de l’I.H.P. Probabilités et statistiques, 31(2):351–377.
  47. [50]Le Donne, E. (2025). Nilpotent Lie Groups. In Metric Lie Groups, volume 306 of Graduate Texts in Mathematics, pages 243–295. Springer.DOI
  48. [51]Levin, D., Lyons, T., and Ni, H. (2013). Learning from the past, predicting the statistics for the future, learning an evolving system. Preprint, arXiv:1309.0260.arxiv.org/abs/1309.0260
  49. [52]Liao, S., Ni, H., Sabaté-Vidales, M., Szpruch, L., Wiese, M., and Xiao, B. (2024). Sig-Wasserstein GANs for conditional time series generation. Mathematical Finance, 34(2):622–670.
  50. [53]Lou, H., Li, S., and Ni, H. (2023). PCF-GAN: generating sequential data via the characteristic function of measures on the path space. In Advances in Neural Information Processing Systems, volume 36, pages 39755–39781.arxiv.org/abs/2305.12511
  51. [54]Lyons, T. (1998). Differential equations driven by rough signals. Revista Matemática Iberoamericana, 14(2):215–310.DOI
  52. [55]Lyons, T., Nejad, S., and Perez Arribas, I. (2019). Numerical Method for Model-Free Pricing of Exotic Derivatives in Discrete Time Using Rough Path Signatures. Applied Mathematical Finance, 26(6):583–597.DOI
  53. [56]Lyons, T. and Ni, H. (2015). Expected signature of Brownian motion up to the first exit time from a bounded domain. Annals of Probability, 43(5):2729–2762.DOI
  54. [57]Lyons, T. and Victoir, N. (2004). Cubature on Wiener Space. Proceedings of the Royal Society of London. Series A: Mathematical, Physical and Engineering Sciences, 460(2041):169–198.DOI
  55. [58]Lyons, T. and Xu, W. (2017). Hyperbolic development and inversion of signature. Journal of Functional Analysis.arxiv.org/abs/1507.00286
  56. [59]Lyons, T. and Xu, W. (2018). Inverting the signature of a path. Journal of the European Mathematical Society, 20(7):1655–1687.arxiv.org/abs/1406.7833
  57. [60]Ni, H. (2012). The Expected Signature of a Stochastic Process. PhD thesis, University of Oxford.DOI
  58. [61]Papavasiliou, A. and Ladroue, C. (2011). Parameter Estimation for Rough Differential Equations. The Annals of Statistics, 39(4):2047–2073.
  59. [62]Perez Arribas, I., Salvi, C., and Szpruch, L. (2021). Sig-SDEs model for quantitative finance. In Proceedings of the First ACM International Conference on AI in Finance, ICAIF ’20, New York, NY, USA.arxiv.org/abs/2006.00218
  60. [63]Popov, V., Yermekova, A., Sadekova, T., Khrapov, A., and Kudinov, M. S. (2025). Improved Sampling Algorithms for Lévy-Itô Diffusion Models. In The Thirteenth International Conference on Learning Representations.
  61. [64]Semnani, P., Guan, V., Robeva, E., and Lee, D. (2025). Path-Dependent SDEs: Solutions and Parameter Estimation. Preprint, arXiv:2505.22646.arxiv.org/abs/2505.22646
  62. [65]Sohl-Dickstein, J., Weiss, E., Maheswaranathan, N., and Ganguli, S. (2015). Deep unsupervised learning using nonequilibrium thermodynamics. In Proceedings of the 32nd International Conference on Machine Learning, volume 37 of Proceedings of Machine Learning Research, pages 2256–2265.
  63. [66]Song, Y. and Ermon, S. (2019). Generative modeling by estimating gradients of the data distribution. In Advances in Neural Information Processing Systems, volume 32.arxiv.org/abs/1907.05600
  64. [67]Song, Y., Sohl-Dickstein, J., Kingma, D. P., Kumar, A., Ermon, S., and Poole, B. (2021). Score-Based Generative Modeling through Stochastic Differential Equations. In International Conference on Learning Representations (ICLR).
  65. [68]Strichartz, R. S. (1986). Sub-Riemannian Geometry. Journal of Differential Geometry, 24(2):221–263.DOI
  66. [69]Tang, W. and Zhao, H. (2025). Score-based diffusion models via stochastic differential equations. Statistical Surveys, 19:28–64.DOI
  67. [70]Vincent, P. (2011). A Connection Between Score Matching and Denoising Autoencoders. Neural Computation, 23(7):1661–1674.DOI
  68. [71]Yoon, E., Park, K., Kim, S., and Lim, S. (2023). Score-based Generative Models with Lévy Processes. In Advances in Neural Information Processing Systems, volume 36.DOI

Paper details

vlnc:W_kveubhde
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Mathematics

Not recorded for this paper

View statement
Looking for related papers…
Contents