mlsanityData to 14 Oct 2026

Deep Signature Transforms

arXiv 1905.08494 ↗First posted 21 May 2019cs.LG stat.MLcited by 75

Not on the map. Why?

Abstract (arXiv)

The signature is an infinite graded sequence of statistics known to characterise a stream of data up to a negligible equivalence class. It is a transform which has previously been treated as a fixed feature transformation, on top of which a model may be built. We propose a novel approach which combines the advantages of the signature transform with modern deep learning frameworks. By learning an augmentation of the stream prior to the signature transform, the terms of the signature may be selected in a data-dependent way. More generally, we describe how the signature transform may be used as a layer anywhere within a neural network. In this context it may be interpreted as a pooling operation. We present the results of empirical experiments to back up the theoretical justification. Code available at https://github.com/patrick-kidger/Deep-Signature-Transforms.

Key facts. Cited by 75 papers in the map (data through 14 Oct 2026). First posted on arXiv 21 May 2019.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2021: 10; 2022: 7; 2023: 18; 2024: 12; 2025: 14; 2026: 14.

75 papers.

Sort:

By area:

  1. 2024Theoretical Foundations of Deep Selective State-Space Modelscited by 74
  2. 2021Efficient and Accurate Gradients for Neural SDEscited by 71
  3. 2024ContiFormer: Continuous-Time Transformer for Irregular Time Series Modelingcited by 63
  4. 2023Vector Quantized Time Series Generation with a Bidirectional Prior Modelcited by 50
  5. 2021Neural Controlled Differential Equations for Online Prediction Taskscited by 46
  6. 2022GT-GAN: General Purpose Time Series Synthesis with Generative Adversarial Networkscited by 44
  7. 2023Non-adversarial training of Neural SDEs with signature kernel scorescited by 37
  8. 2021Higher Order Kernel Mean Embeddings to Capture Filtrations of Stochastic Processescited by 27
  9. 2023Neural signature kernels as infinite-width-depth-limits of controlled ResNetscited by 25
  10. 2023Recent Developments in Machine Learning Methods for Stochastic Control and Gamescited by 24
  11. 2021Framing RNN as a kernel method: A neural ODE approachcited by 23
  12. 2025Structured Linear CDEs: Maximally Expressive and Parallel-in-Time Sequence Modelscited by 18
38 more
  1. 2024Rough Transformers: Lightweight and Continuous Time Series Modelling through Signature Patchingcited by 16
  2. 2023PCF-GAN: generating sequential data via the characteristic function of measures on the path spacecited by 15
  3. 2021Signatured Deep Fictitious Play for Mean Field Games with Common Noisecited by 12
  4. 2021Approximate Bayesian Computation with Path Signaturescited by 12
  5. 2023New directions in the applications of rough path theorycited by 12
  6. 2021Logsig-RNN: a novel network for robust and efficient skeleton-based action recognitioncited by 11
  7. 2022Path Development Network with Finite-dimensional Lie Group Representationcited by 8
  8. 2021Feature Engineering with Regularity Structurescited by 7
  9. 2022Optimal Estimation of Generic Dynamics by Path-Dependent Neural Jump ODEscited by 7
  10. 2023SigFormer: Signature Transformers for Deep Hedgingcited by 7
  11. 2024Dynamical Survival Analysis with Controlled Latent Statescited by 7
  12. 2022Deep Signature Algorithm for Multi-dimensional Path-Dependent Optionscited by 6
  13. 2024Sparse Signature Coefficient Recovery via Kernelscited by 6
  14. 2025pySigLib -- Fast Signature-Based Computations on CPU and GPUcited by 6
  15. 2023Risk of Transfer Learning and its Applications in Financecited by 5
  16. 2021Deep Signature FBSDE Algorithmcited by 4
  17. 2023G-Signatures: Global Graph Propagation With Randomized Signaturescited by 4
  18. 2024Novelty Detection on Radio Astronomy Data using Signaturescited by 4
  19. 2025ParallelFlow: Parallelizing Linear Transformers via Flow Discretizationcited by 4
  20. 2023Learning the Dynamics of Sparsely Observed Interacting Systemscited by 3
  21. 2024Tensor learning with orthogonal, Lorentz, and symplectic symmetriescited by 3
  22. 2025Hedging with memory: shallow and deep learning with signaturescited by 3
  23. 2022ImageSig: A signature transform for ultra-lightweight image recognitioncited by 2
  24. 2023Sig-Networks Toolkit: Signature Networks for Longitudinal Language Modellingcited by 2
  25. 2024Deep learning the Hurst parameter of linear fractional processes and assessing its reliabilitycited by 2
  26. 2024Parameter Estimation of Long Memory Stochastic Processes with Deep Neural Networkscited by 2
  27. 2025Signature Kernel Scoring Rule: A Spatio-Temporal Diagnostic for Probabilistic Weather Forecastingcited by 2
  28. 2025Global universal approximation with Brownian signaturescited by 2
  29. 2026Minimax Lower Bounds of Kernel Discrepancy Estimation: MMD, HSIC, KSDcited by 2
  30. 2021Modelling Paralinguistic Properties in Conversational Speech to Detect Bipolar Disorder and Borderline Personality Disordercited by 1
  31. 2023FRUITS: Feature Extraction Using Iterated Sums for Time Series Classificationcited by 1
  32. 2024Transportation Marketplace Rate Forecast Using Signature Transformcited by 1
  33. 2025Expected Signature Kernels for Rough Pathscited by 1
  34. 2025Local regression on path spaces with signature metricscited by 1
  35. 2025Path Signatures Enable Model-Free Mapping of RNA Modificationscited by 1
  36. 2022Signature and Log-signature for the Study of Empirical Distributions Generated with GANscited by 0
  37. 2022Learning with Signaturescited by 0
  38. 2023Directed Chain Generative Adversarial Networkscited by 0

50 most cited of 75 All 75 (JSON)