Published articles

  • Functional linear regression with truncated signatures. Fermanian, A. (2022)
    Journal of Multivariate Analysis, Vol. 192, 105031
    pdf code

  • Framing RNN as a kernel method: A neural ODE approach. Fermanian, A., Marion. P., Vert, J.P., Biau, G. (2021)
    in Advances in Neural Information Processing Systems, ed. Ranzato, M., Beygelzimer, A., Dauphin, Y., Liang, P.S. and Wortman Vaughan, J., Vol. 34, pp. 3121-3134, Curran Associates, Inc. (oral presentation)
    pdf code

  • Embedding and learning with signatures. Fermanian, A. (2021)
    Computational Statistics & Data Analysis, Vol. 157, 107148
    pdf code

Preprints

  • Scaling ResNets in the Large-depth Regime. Marion P., Fermanian, A., Biau G., Vert, J.P. (2022)
    pdf code

  • A Generalised Signature Method for Multivariate Time Series Feature Extraction. Morrill, J., Fermanian, A., Kidger, P., Lyons, T. (2020)
    pdf code

PhD dissertation

I have defended my PhD thesis in October 2021, you can find my dissertation, entitled Learning time-dependent data with the signature transform here.

Conference proceedings and miscellaneous

  • Learning with signatures. Biau, G. and Fermanian, A. (2020)
    in Functional and High-Dimensional Statistics and Related Fields, pp. 19-26, Springer, Cham.
    pdf

  • Master thesis.(2018)
    pdf