Nikolas Nüsken

Nikolas Nüsken

Lecturer (Assistant Professor) Mathematical Data Science

King's College London

About me.

I am a lecturer (assistant professor) in the Maths Department at King’s College London.

My research interests lie at the interface of computational statistics, machine learning, and the mathematics of stochastic dynamics and interacting particle systems.

Interests

  • Computational Bayesian inference
  • PDEs and stochastic analysis for interacting particle systems
  • Optimal control and optimal transport
  • Kernel methods

Education

  • PhD in Mathematics, 2018

    Imperial College London

  • MSc in Mathematics, 2014

    Leibniz Universität Hannover

  • BSc in Physics, 2011

    Leibniz Universität Hannover

Publications and Preprints

R. Polzin, I. Klebanov, N. Nüsken, P. Koltai (2023). Nonnegative matrix factorization for coherent set identification by direct low rank maximum likelihood estimation. (submitted).

arxiv

L. Richter, L. Sallandt, N. Nüsken (2023). From continuous-time formulations to discretization schemes: tensor trains and robust regression for BSDEs and parabolic PDEs. (submitted).

arxiv

F. Vargas, N. Nüsken (2023). Transport, Variational Inference and Diffusions: with Applications to Annealed Flows and Schrödinger Bridges. In ICML Workshop.

PDF arxiv

A. Duncan, N. Nüsken, L. Szpruch (2023). On the geometry of Stein variational gradient descent. In JMLR.

arxiv

N. Nüsken, D.R.M. Renger (2022). Stein variational gradient descent: many-particle and long-time asymptotics. In FoDS.

PDF arxiv

N. Nüsken, L. Richter (2022). Interpolating between BSDEs and PINNs – deep learning for elliptic and parabolic boundary value problems. In JML.

PDF arxiv

F. Vargas, A. Ovsianas, D. Fernandes, M. Girolami, N. D. Lawrence, N. Nüsken (2022). Bayesian Learning via Neural Schrödinger-Föllmer Flows. In Statistics and Computing.

PDF arxiv

C. Andrieu, A. Durmus, N. Nüsken, J. Roussel (2021). Hypocoercivity of Piecewise Deterministic Markov Process-Monte Carlo. In AAP.

PDF arxiv

M. Coghi, T. Nilssen, N. Nüsken, S. Reich (2021). Rough McKean-Vlasov dynamics for robust ensemble Kalman filtering. (submitted).

arxiv

N. Nüsken, L. Richter (2021). Solving high-dimensional Hamilton-Jacobi-Bellman PDEs using neural networks: perspectives from the theory of controlled diffusions and measures on path space. In PDEA.

PDF arxiv

L. Richter, L. Sallandt, N. Nüsken (2021). Solving high-dimensional parabolic PDEs using the tensor train format. In ICML.

PDF arxiv

L. Richter, A. Boustati, N. Nüsken, A. Akyildiz, F.J.R. Ruiz (2020). VarGrad: A Low-Variance Gradient Estimator for Variational Inference. In NeurIPS.

PDF arxiv

A. Garbuno-Inigo, N. Nüsken, S. Reich (2020). Affine Invariant Interacting Langevin Dynamics for Bayesian Inference. In SIAM J. Appl. Dyn. Syst..

PDF arxiv

N. Nüsken, S. Reich (2019). Note on Interacting Langevin Diffusions: Gradient Structure and Ensemble Kalman Sampler by Garbuno-Inigo, Hoffmann, Li and Stuart. (unpublished).

arxiv

N. Nüsken, S. Reich, P. Rozdeba (2019). State and Parameter Estimation from Observed Signal Increments. In Entropy.

PDF arxiv

N. Nüsken, G.A. Pavliotis (2019). Constructing sampling schemes via coupling: Markov semigroups and optimal transport. In SIAM UQ.

PDF arxiv

N. Nüsken (2018). Topics in sampling schemes based on Markov processes. PhD thesis.

PDF

A. Duncan, N. Nüsken, G.A. Pavliotis (2017). Using Perturbed Underdamped Langevin Dynamics to Efficiently Sample from Probability Distributions. In J. Stat. Phys..

PDF arxiv

Contact