Publications

Leveraging Continuous Time to Understand Momentum When Training Diagonal Linear Networks

H. G. Papazov; S. Pesme; N. Flammarion 

2024-03-10. Proceedings of the 27th International Conference on Artificial Intelligence and Statistics (AIS- TATS) 2024, , Valencia, Spain, May 2-4, 2024.

Deep Learning Theory Through the Lens of Diagonal Linear Networks

S. W. Pesme / N. H. B. Flammarion (Dir.)  

Lausanne, EPFL, 2024. 

Understanding generalization and robustness in modern deep learning

M. Andriushchenko / N. H. B. Flammarion (Dir.)  

Lausanne, EPFL, 2024. 

Scalable constrained optimization

M-L. Vladarean / N. H. B. Flammarion (Dir.)  

Lausanne, EPFL, 2024. 

Saddle-to-Saddle Dynamics in Diagonal Linear Networks

S. Pesme; N. Flammarion 

2023-04-02. 37th Conference on Neural Information Processing Systems (NeurIPS 2023), New Orleans, LA, United States, December 10-16, 2023.

Model agnostic methods meta-learn despite misspecifications

O. K. Yüksel; E. Boursier; N. Flammarion 

2023-03-03

Penalising the biases in norm regularisation enforces sparsity

E. Boursier; N. Flammarion 

2023-03-03

(S)GD over Diagonal Linear Networks: Implicit Regularisation, Large Stepsizes and Edge of Stability

M. Even; S. Pesme; S. Gunasekar; N. Flammarion 

2023-02-17. 37th Conference on Neural Information Processing Systems (NeurIPS 2023), New Orleans, LA, United States, December 10-16, 2023.

Accelerated SGD for Non-Strongly-Convex Least Squares

A. V. Varre; N. Flammarion 

2022-03-03

An Efficient Sampling Algorithm for Non-smooth Composite Potentials

W. Mou; N. Flammarion; M. J. Wainwright; P. L. Bartlett 

Journal Of Machine Learning Research. 2022-01-01. Vol. 23.

Towards Understanding Sharpness-Aware Minimization

M. Andriushchenko; N. Flammarion 

2022-01-01. 38th International Conference on Machine Learning (ICML), Baltimore, MD, Jul 17-23, 2022. p. 639-668.

Sparse-RS: A Versatile Framework for Query-Efficient Sparse Black-Box Adversarial Attacks

F. Croce; M. Andriushchenko; N. D. Singh; N. Flammarion; M. Hein 

2022-01-01. 36th AAAI Conference on Artificial Intelligence / 34th Conference on Innovative Applications of Artificial Intelligence / 12th Symposium on Educational Advances in Artificial Intelligence, ELECTR NETWORK, Feb 22-Mar 01, 2022. p. 6437-6445. DOI : 10.1609/aaai.v36i6.20595.

Gradient flow dynamics of shallow ReLU networks for square loss and orthogonal inputs

E. Boursier; L. Pillaud-Vivien; N. Flammarion 

2022

Utility/privacy trade-off as regularized optimal transport

E. Boursier; V. Perchet 

Mathematical Programming. 2022-04-22. DOI : 10.1007/s10107-022-01811-w.

Trace norm regularization for multi-task learning with scarce data

E. Boursier; M. Konobeev; N. Flammarion 

2022

Improved bounds for discretization of Langevin diffusions: Near-optimal rates without convexity

W. Mou; N. Flammarion; M. J. Wainwright; P. L. Bartlett 

Bernoulli. 2022-08-01. Vol. 28, num. 3, p. 1577-1601. DOI : 10.3150/21-BEJ1343.

Last iterate convergence of SGD for Least-Squares in the Interpolation regime

A. Varre; L. Pillaud-Vivien; N. Flammarion 

2021-09-28

Is there an analog of Nesterov acceleration for gradient-based MCMC?

Y-A. Ma; N. S. Chatterji; X. Cheng; N. Flammarion; P. L. Bartlett et al. 

Bernoulli. 2021-08-01. Vol. 27, num. 3, p. 1942-1992. DOI : 10.3150/20-BEJ1297.

Implicit Bias of SGD for Diagonal Linear Networks: a Provable Benefit of Stochasticity

S. Pesme; L. Pillaud-Vivien; N. Flammarion 

2021-06-16. 35th Conference on Neural Information Processing Systems (NeurIPS 2021), Virtual Conference, December 6-14, 2021.

On the effectiveness of adversarial training against common corruptions

K. Kireev; M. Andriushchenko; N. Flammarion 

2021-03-03

A Continuized View on Nesterov Acceleration

R. Berthier; F. Bach; N. Flammarion; P. Gaillard; A. Taylor 

2021

On the Stability of Fine-tuning BERT: Misconceptions, Explanations, and Strong Baselines

M. Mosbach; M. Andriushchenko; D. Klakow 

2021. 9th International Conference on Learning Representations, Virtual, May 4-8, 2021.

RobustBench: a standardized adversarial robustness benchmark

F. Croce; M. Andriushchenko; V. Sehwag; N. Flammarion; M. Chiang et al. 

2020-10-19

Understanding and Improving Fast Adversarial Training

M. Andriushchenko; N. Flammarion 

2020-07-06. Advances In Neural Information Processing Systems 33 (NeurIPS 2020), [Online], December 2020.

Sparse-RS: a versatile framework for query-efficient sparse black-box adversarial attacks

F. Croce; M. Andriushchenko; N. Singh; N. Flammarion; M. Hein 

2020-06-23

Online Robust Regression via SGD on the l1 loss

S. Pesme; N. Flammarion 

2020. Neurips 2020.

On Convergence-Diagnostic based Step Sizes for Stochastic Gradient Descent

S. Pesme; A. D. K. Dieuleveut; N. Flammarion 

2020. 37th International Conference on Machine Learning (ICLM 2020), [Online event], July 12-18, 2020.

Square Attack: a query-efficient black-box adversarial attack via random search

M. Andriushchenko; F. Croce; N. Flammarion; M. Hein 

2020-08-28. European Conference on Computer Vision (ECCV 2020), [Online], August 23-28, 2020. p. 484–501. DOI : 10.1007/978-3-030-58592-1_29.

Provably Robust Boosted Decision Stumps and Trees against Adversarial Attacks

M. Andriushchenko; M. Hein 

arXiv. 2019-10-30.  p. 1906.03526 [cs, stat].

Escaping from saddle points on Riemannian manifolds

Y. Sun; N. Flammarion; M. Fazel 

2019. 33rd Conference on Neural Information Processing Systems (NeurIPS), Vancouver, CANADA, Dec 08-14, 2019.

Is There an Analog of Nesterov Acceleration for MCMC?

Y-A. Ma; N. Chatterji; X. Cheng; N. Flammarion; P. L. Bartlett et al. 

arXiv. 2019. 

Fast Mean Estimation with Sub-Gaussian Rates

Y. Cherapanamjeri; N. Flammarion; P. L. Bartlett 

2019. 

Optimal rates of statistical seriation

N. Flammarion; C. Mao; P. Rigollet 

Bernoulli. 2019. Vol. 25, num. 1, p. 623-653. DOI : 10.3150/17-BEJ1000.

Sampling can be faster than optimization

Y-A. Ma; Y. Chen; C. Jin; N. Flammarion; M. I. Jordan 

Proceedings of the National Academy of Sciences. 2019. Vol. 116, num. 42, p. 20881-20885. DOI : 10.1073/pnas.1820003116.

Gen-Oja: A Simple and Efficient Algorithm for Streaming Generalized Eigenvector Computation

K. Bhatia; A. Pacchiano; N. Flammarion; P. L. Bartlett; M. I. Jordan 

2018. Neural Information Processing Systems Conference NIPS 2018.

Averaging Stochastic Gradient Descent on Riemannian Manifolds

N. Tripuraneni; N. Flammarion; F. Bach; M. I. Jordan 

2018. 

On the theory of variance reduction for stochastic gradient monte carlo

N. Chatterji; N. Flammarion; Y-A. Ma; P. Bartlett; M. I. Jordan 

2018. 

Stochastic Composite Least-Squares Regression with Convergence Rate O(1/n)

N. Flammarion; F. Bach 

2017. 

Robust Discriminative Clustering with Sparse Regularizers

N. Flammarion; B. Palaniappan; F. Bach 

Journal of Machine Learning Research. 2017. Vol. 18.

Harder, Better, Faster, Stronger Convergence Rates for Least-Squares Regression

A. Dieuleveut; N. Flammarion; F. Bach 

Journal of Machine Learning Research. 2017. Vol. 18.

From averaging to acceleration, there is only a step-size

N. Flammarion; F. Bach 

2015. Conference on Learning Theory (COLT).