Patrick Rebeschini

59 papers A* 15A 4Misc 1Journal 37Unranked 1

Year	Rank	Type	Title / Venue / Authors
2026	J	`jnl`	On-Average Stability of Multipass Preconditioned SGD and Effective Dimension. CoRR Simon Vary, Tyler Farghly, Ilja Kuzborskij, Patrick Rebeschini
2025	A	`conf`	Black-Box Uniform Stability for Non-Euclidean Empirical Risk Minimization. AISTATS Simon Vary, David Martínez-Rubio, Patrick Rebeschini
2025	J	`jnl`	Early-Stopped Mirror Descent for Linear Regression over Convex Bodies. CoRR Tobias Wegel, Gil Kur, Patrick Rebeschini
2025	J	`jnl`	Implicit Regularisation in Diffusion Models: An Algorithm-Dependent Generalisation Analysis. CoRR Tyler Farghly, Patrick Rebeschini, George Deligiannidis, Arnaud Doucet
2025	A*	`conf`	Learning mirror maps in policy mirror descent. ICLR Carlo Alfano, Sebastian Rene Towers, Silvia Sapora, Chris Lu, Patrick Rebeschini
2025	J	`jnl`	Non-stationary Bandit Convex Optimization: A Comprehensive Study. CoRR Xiaoqi Liu, Dorian Baudry, Julian Zimmert, Patrick Rebeschini, Arya Akhavan
2025	J	`jnl`	On the necessity of adaptive regularisation:Optimal anytime online learning on ℓ CoRR Emmeran Johnson, David Martínez-Rubio, Ciara Pike-Burke, Patrick Rebeschini
2025	A	`conf`	Robust Gradient Descent for Phase Retrieval. AISTATS Alex Buna, Patrick Rebeschini
2025	J	`jnl`	Self-Concordant Perturbations for Linear Bandits. CoRR Lucas Lévy, Jean-Lou Valeau, Arya Akhavan, Patrick Rebeschini
2025	J	`jnl`	Stochastic Shortest Path with Sparse Adversarial Costs. CoRR Emmeran Johnson, Alberto Rumi, Ciara Pike-Burke, Patrick Rebeschini
2024	J	`jnl`	Black-Box Uniform Stability for Non-Euclidean Empirical Risk Minimization. CoRR Simon Vary, David Martínez-Rubio, Patrick Rebeschini
2024	J	`jnl`	Differentiable Cost-Parameterized Monge Map Estimators. CoRR Samuel Howard, George Deligiannidis, Patrick Rebeschini, James Thornton
2024	J	`jnl`	Exponential Tail Local Rademacher Complexity Risk Bounds Without the Bernstein Condition. J. Mach. Learn. Res. Varun Kanade, Patrick Rebeschini, Tomas Vaskevicius
2024	A	`conf`	Generalization Bounds for Label Noise Stochastic Gradient Descent. AISTATS Jung Eun Huh, Patrick Rebeschini
2024	J	`jnl`	Learning Loss Landscapes in Preference Optimization. CoRR Carlo Alfano, Silvia Sapora, Jakob Nicolaus Foerster, Patrick Rebeschini, Yee Whye Teh
2024	J	`jnl`	Meta-learning the mirror map in policy mirror descent. CoRR Carlo Alfano, Sebastian Towers, Silvia Sapora, Chris Lu, Patrick Rebeschini
2024	J	`jnl`	Robust Gradient Descent for Phase Retrieval. CoRR Alex Buna, Patrick Rebeschini
2024	A*	`conf`	Sample-Efficiency in Multi-Batch Reinforcement Learning: The Need for Dimension-Dependent Adaptivity. ICLR Emmeran Johnson, Ciara Pike-Burke, Patrick Rebeschini
2023	A*	`conf`	A Novel Framework for Policy Mirror Descent with General Parameterization and Linear Convergence. NeurIPS Carlo Alfano, Rui Yuan, Patrick Rebeschini
2023	J	`jnl`	A Novel Framework for Policy Mirror Descent with General Parametrization and Linear Convergence. CoRR Carlo Alfano, Rui Yuan, Patrick Rebeschini
2023	J	`jnl`	Generalization Bounds for Label Noise Stochastic Gradient Descent. CoRR Jung Eun Huh, Patrick Rebeschini
2023	A*	`conf`	Optimal Convergence Rate for Exact Policy Mirror Descent in Discounted Markov Decision Processes. NeurIPS Emmeran Johnson, Ciara Pike-Burke, Patrick Rebeschini
2023	J	`jnl`	Optimal Convergence Rate for Exact Policy Mirror Descent in Discounted Markov Decision Processes. CoRR Emmeran Johnson, Ciara Pike-Burke, Patrick Rebeschini
2023	J	`jnl`	Sample-Efficiency in Multi-Batch Reinforcement Learning: The Need for Dimension-Dependent Adaptivity. CoRR Emmeran Johnson, Ciara Pike-Burke, Patrick Rebeschini
2022	J	`jnl`	Exponential Tail Local Rademacher Complexity Risk Bounds Without the Bernstein Condition. CoRR Varun Kanade, Patrick Rebeschini, Tomas Vaskevicius
2022	J	`jnl`	Linear Convergence for Natural Policy Gradient with Log-linear Policy Parametrization. CoRR Carlo Alfano, Patrick Rebeschini
2021	J	`jnl`	Comparing Classes of Estimators: When does Gradient Descent Beat Ridge Regression in Linear Models? CoRR Dominic Richards, Edgar Dobriban, Patrick Rebeschini
2021	J	`jnl`	Dimension-Free Rates for Natural Policy Gradient in Multi-Agent Reinforcement Learning. CoRR Carlo Alfano, Patrick Rebeschini
2021	A*	`conf`	Distributed Machine Learning with Sparse Heterogeneous Data. NeurIPS Dominic Richards, Sahand Negahban, Patrick Rebeschini
2021	A	`conf`	Hadamard Wirtinger Flow for Sparse Phase Retrieval. AISTATS Fan Wu, Patrick Rebeschini
2021	A*	`conf`	Implicit Regularization in Matrix Sensing via Mirror Descent. NeurIPS Fan Wu, Patrick Rebeschini
2021	J	`jnl`	Implicit Regularization in Matrix Sensing via Mirror Descent. CoRR Fan Wu, Patrick Rebeschini
2021	J	`jnl`	Nearly Minimax-Optimal Rates for Noisy Sparse Phase Retrieval via Early-Stopped Mirror Descent. CoRR Fan Wu, Patrick Rebeschini
2021	J	`jnl`	On Optimal Interpolation In Linear Regression. CoRR Eduard Oravkin, Patrick Rebeschini
2021	A*	`conf`	On Optimal Interpolation in Linear Regression. NeurIPS Eduard Oravkin, Patrick Rebeschini
2021	A*	`conf`	Time-independent Generalization Bounds for SGLD in Non-convex Settings. NeurIPS Tyler Farghly, Patrick Rebeschini
2021	J	`jnl`	Time-independent Generalization Bounds for SGLD in Non-convex Settings. CoRR Tyler Farghly, Patrick Rebeschini
2020	A*	`conf`	A Continuous-Time Mirror Descent Approach to Sparse Phase Retrieval. NeurIPS Fan Wu, Patrick Rebeschini
2020	J	`jnl`	A Continuous-Time Mirror Descent Approach to Sparse Phase Retrieval. CoRR Fan Wu, Patrick Rebeschini
2020	A*	`conf`	Decentralised Learning with Random Features and Distributed Gradient Descent. ICML Dominic Richards, Patrick Rebeschini, Lorenzo Rosasco
2020	J	`jnl`	Decentralised Learning with Random Features and Distributed Gradient Descent. CoRR Dominic Richards, Patrick Rebeschini, Lorenzo Rosasco
2020	J	`jnl`	Graph-Dependent Implicit Regularisation for Distributed Stochastic Subgradient Descent. J. Mach. Learn. Res. Dominic Richards, Patrick Rebeschini
2020	J	`jnl`	Hadamard Wirtinger Flow for Sparse Phase Retrieval. CoRR Fan Wu, Patrick Rebeschini
2020	J	`jnl`	The Statistical Complexity of Early Stopped Mirror Descent. CoRR Tomas Vaskevicius, Varun Kanade, Patrick Rebeschini
2020	A*	`conf`	The Statistical Complexity of Early-Stopped Mirror Descent. NeurIPS Tomas Vaskevicius, Varun Kanade, Patrick Rebeschini
2019	J	`jnl`	A New Approach to Laplacian Solvers and Flow Problems. J. Mach. Learn. Res. Patrick Rebeschini, Sekhar Tatikonda
2019	A*	`conf`	Decentralized Cooperative Stochastic Bandits. NeurIPS David Martínez-Rubio, Varun Kanade, Patrick Rebeschini
2019	A*	`conf`	Implicit Regularization for Optimal Sparse Recovery. NeurIPS Tomas Vaskevicius, Varun Kanade, Patrick Rebeschini
2019	J	`jnl`	Implicit Regularization for Optimal Sparse Recovery. CoRR Tomas Vaskevicius, Varun Kanade, Patrick Rebeschini
2019	J	`jnl`	Locality in Network Optimization. IEEE Trans. Control. Netw. Syst. Patrick Rebeschini, Sekhar Tatikonda
2019	A*	`conf`	Optimal Statistical Rates for Decentralised Non-Parametric Regression with Linear Speed-Up. NeurIPS Dominic Richards, Patrick Rebeschini
2019	J	`jnl`	Optimal Statistical Rates for Decentralised Non-Parametric Regression with Linear Speed-Up. CoRR Dominic Richards, Patrick Rebeschini
2018	J	`jnl`	Decentralized Cooperative Stochastic Multi-armed Bandits. CoRR David Martínez-Rubio, Varun Kanade, Patrick Rebeschini
2018	J	`jnl`	Graph-Dependent Implicit Regularisation for Distributed Stochastic Subgradient Descent. CoRR Dominic Richards, Patrick Rebeschini
2017	—	`conf`	Accelerated consensus via Min-Sum Splitting. NIPS Patrick Rebeschini, Sekhar Tatikonda
2016	J	`jnl`	A new approach to Laplacian solvers and flow problems. CoRR Patrick Rebeschini, Sekhar Tatikonda
2016	Misc	`conf`	Decay of correlation in network flow problems. CISS Patrick Rebeschini, Sekhar Tatikonda
2015	A*	`conf`	Fast Mixing for Discrete Point Processes. COLT Patrick Rebeschini, Amin Karbasi
2014	—	`—`	Nonlinear Filtering in High Dimension Patrick Rebeschini