Publications

Journals and international conferences

• Benjamin Dupuis, Paul Viallard, George Deligiannidis, Umut Simsekli, “Uniform Generalization Bounds on Data-Dependent Hypothesis Sets via PAC-Bayesian Theory on Random Sets”, Jounral of Machine Learning Research, JMLR, 2024

• Rayna Andreeva, Benjamin Dupuis, Rik Sarkar, Tolga Birdal, Umut Simsekli, “Topological Generalization Bounds for Discrete-Time Stochastic Optimization Algorithms”, Advances in Neural Information Processing Systems Conference NeurIPS, New Orleans, USA, 2024

• Benjamin Dupuis, Umut Simsekli “Generalization Bounds for Heavy-Tailed SDEs through the Fractional Fokker-Planck Equation”, International Conference on Machine Learning (ICML), Vienna, Austria, 2024

• Benjamin Dupuis, George Deligiannidis, Umut Simsekli “Generalization Bounds with Data-dependent Fractal Dimensions”, International Conference on Machine Learning (ICML), Honolulu, Hawaii, USA, 2023

• Benjamin Dupuis, Arthur Jacot, “DNN-based topology optimisation: Spatial invariance and neural tangent kernel”, Advances in Neural Information Processing Systems Conference (NeurIPS), 2021

Workshops

• Benjamin Dupuis, Paul Viallard, “From Mutual Information to Expected Dynamics: New Generalization Bounds for Heavy-Tailed SGD”, Heavy Tails in Machine Learning Workshop at Neurips 2023, 2023

Preprints

• Benjamin Dupuis, Maxime Haddouche, George Deligiannidis, Umut Simsekli, “Understanding the Generalization Error of Markov algorithms through Poissonization”

Master thesis

• Benjamin Dupuis Generalization Bounds with Data-dependent Fractal Dimensions, 2023