Research Interests

• High Dimensional Statistics
• Spectral Methods
• Operator Learning
• Machine Learning for Science
• Nonparametric Statistics
• Bandits
• Unsupervised Learning

Click here for recent research projects

Submitted Papers

• A conversion theorem and minimax optimality for continuum contextual bandits
  Arya Akhavan, Karim Lounici, Massimiliano Pontil, and Alexandre B. Tsybakov, 2024.
  [arXiv].


• Multi-Source and Test-Time Domain Adaptation on Multivariate Signals using Spatio-Temporal Monge Alignment
  Théo Gnassounou, Antoine Collas, Rémi Flamary, Karim Lounici, and Alexandre Gramfort, 2024.
  [arXiv].


• Equivariant Representation Learning for Symmetry-Aware Inference with Guarantees
  Danie Ordoñez-Apraez, Alek Frohlick, Vladimir Kostic, Karim Lounici, Vivien Brandt, and Massimiliano Pontil, 2025.
  [arXiv].

Accepted/Published Papers

• Meta representation learning with contextual linear bandits
  Leonardo Cella, Karim Lounici, and Massimiliano Pontil.
  To appear in Statistical Science.
  [arXiv].


• Laplace transform based low-complexity learning of continuous Markov semigroups
  Vladimir R. Kostic, Karim Lounici, Hélène Halconruy, Timothée Devergne, Pietro Novelli, and Massimiliano Pontil.
  In ICML 2025.


• Learning the infinitesimal generator of stochastic diffusion processes
  Vladimir Kostic, Hélène Halconruy, Timothée Devergne, Karim Lounici, and Massimiliano Pontil.
  In NeurIPS 2024.


• Consistent long-term forecasting of ergodic dynamical systems
  Vladimir Kostic, Prune Inzerili, Karim Lounici, Pietro Novelli, and Massimiliano Pontil.
  In ICML 2024.


• Neural Conditional Probability for Uncertainty Quantification
  Vladimir Kostic, Grégoire Pacreau, Giacomo Turri, Pietro Novelli, Karim Lounici, and Massimiliano Pontil.
  In NeurIPS 2024.


• Learning invariant representations of time-homogeneous stochastic dynamical systems
  Vladimir R. Kostic, Pietro Novelli, Riccardo Grazzi, Karim Lounici, and Massimiliano Pontil.
  In ICLR 2024.


• Multi-task representation learning with stochastic linear bandits
  Leonardo Cella, Karim Lounici, Grégoire Pacreau, and Massimiliano Pontil.
  In AISTATS 2023.


• Sharp spectral rates for Koopman operator learning
  Vladimir Kostic, Karim Lounici, Pietro Novelli, and Massimiliano Pontil.
  In NeurIPS 2023 (Spotlight paper).


• Robust covariance estimation with missing values and cell-wise contamination
  Grégoire Pacreau and Karim Lounici.
  In NeurIPS 2023.


• Sliding window Strategy for convolutional spike sorting with Lasso
  Laurent Dragoni, Rémi Flamary, Karim Lounici, and Patricia Reynaud-Bouret.
  In Acta Applicandae Mathematicae, 179(7), 2022.


• Adaptive sup-norm estimation of the Wigner function in noisy quantum homodyne tomography
  Karim Lounici, Katia Meziani, and Gabriel Peyré.
  In Annals of Statistics, 46(3):1318–1351, 2018.


• Robust matrix completion
  Olga Klopp, Karim Lounici, and Alexandre B. Tsybakov.
  In Probability Theory and Related Fields, 169:523–564, 2017.


• Concentration inequalities and moment bounds for sample covariance operators
  Vladimir Koltchinskii and Karim Lounici.
  In Bernoulli, 2017.


• New asymptotic results in principal component analysis
  Vladimir Koltchinskii and Karim Lounici.
  In Sankhya A, 79:254–297, 2017.


• Normal approximation and concentration of spectral projectors of sample covariance
  Vladimir Koltchinskii and Karim Lounici.
  In Annals of Statistics, 45(1):121–157, 2017.


• Asymptotics and concentration bounds for bilinear forms of spectral projectors of sample covariance
  Vladimir Koltchinskii and Karim Lounici.
  In Ann. Inst. Henri Poincaré, 52(4):1976–2013, 2016.


• Estimation of low-rank covariance function
  Vladimir Koltchinskii, Karim Lounici, and Alexandre B. Tsybakov.
  In Stochastic Processes and their Applications, 126(12):3952–3967, 2016.


• Estimation and variable selection with exponential weights
  Ery Arias-Castro and Karim Lounici.
  In Electronic Journal of Statistics, 8(1):328–354, 2014.


• High-dimensional covariance matrix estimation with missing observations
  Karim Lounici.
  In Bernoulli, 20(3):1029–1058, 2014.


• Sparse principal component analysis with missing observations
  Karim Lounici.
  In High Dimensional Probability VI, pages 327–356, 2013.


• PAC-Bayesian bounds for sparse regression estimation with exponential weights
  Pierre Alquier and Karim Lounici.
  In Electronic Journal of Statistics, 5:127–145, 2011.


• Nuclear-norm penalization and optimal rates for noisy low-rank matrix completion
  Vladimir Koltchinskii, Karim Lounici, and Alexandre B. Tsybakov.
  In Annals of Statistics, 39(5):2302–2329, 2011.


• Global uniform risk bounds for wavelet deconvolution estimators
  Karim Lounici and Richard Nickl.
  In Annals of Statistics, 39(1):201–231, 2011.


• Oracle inequalities and optimal inference under group sparsity
  Karim Lounici, Massimiliano Pontil, Sara Van De Geer, and Alexandre B. Tsybakov.
  In Annals of Statistics, 39(4):2164–2204, 2011.


• Taking advantage of sparsity in multi-task learning
  Karim Lounici, Massimiliano Pontil, Alexandre B. Tsybakov, and Sara Van De Geer.
  In Proceedings of the 22nd Conference on Information Theory, 2009.


• Sup-norm convergence rate and sign concentration property of Lasso and Dantzig estimators
  Karim Lounici.
  In Electronic Journal of Statistics, 2:90–102, 2008.


• Generalized mirror averaging and D-convex aggregation
  Karim Lounici.
  In Mathematical Methods of Statistics, 16:246–259, 2007.