Mert Gürbüzbalaban

93 papers A* 15A 4B 2Misc 1Journal 66Unranked 5

Year	Rank	Type	Title / Venue / Authors
2026	J	`jnl`	Mean-semideviation-based distributionally robust learning with weakly convex losses: convergence rates and finite-sample guarantees. Math. Program. Landi Zhu, Mert Gürbüzbalaban, Andrzej Ruszczynski
2025	J	`jnl`	Accelerated Gradient Methods with Biased Gradient Estimates: Risk Sensitivity, High-Probability Guarantees, and Large Deviation Bounds. CoRR Mert Gürbüzbalaban, Yasa Syed, Necdet Serhat Aybat
2025	J	`jnl`	Algorithmic Stability of Stochastic Gradient Descent with Momentum under Heavy-Tailed Noise. CoRR Thanh Dang, Melih Barsbey, A K. M. Rokonuzzaman Sonet, Mert Gürbüzbalaban, Umut Simsekli, Lingjiong Zhu
2025	J	`jnl`	Anchored Langevin Algorithms. CoRR Mert Gürbüzbalaban, Hoang M. Nguyen, Xicheng Zhang, Lingjiong Zhu
2025	J	`jnl`	DIGing-SGLD: Decentralized and Scalable Langevin Sampling over Time-Varying Networks. CoRR Waheed U. Bajwa, Mert Gürbüzbalaban, Mustafa Ali Kutbay, Lingjiong Zhu, Muhammad Zulqarnain
2025	J	`jnl`	Entropic risk-averse generalized momentum methods. Optim. Methods Softw. Bugra Can, Mert Gürbüzbalaban
2025	J	`jnl`	High-Order Langevin Monte Carlo Algorithms. CoRR Thanh Dang, Mert Gürbüzbalaban, Mohammad Rafiqul Islam, Nian Yao, Lingjiong Zhu
2025	J	`jnl`	RESIST: Resilient Decentralized Learning Using Consensus Gradient Descent. CoRR Cheng Fang, Rishabh Dixit, Waheed U. Bajwa, Mert Gürbüzbalaban
2025	J	`jnl`	Rényi Differential Privacy for Heavy-Tailed SDEs via Fractional Poincaré Inequalities. CoRR Benjamin Dupuis, Mert Gürbüzbalaban, Umut Simsekli, Jian Wang, Sinan Yildirim, Lingjiong Zhu
2024	J	`jnl`	Differential Privacy of Noisy (S)GD under Heavy-Tailed Perturbations. CoRR Umut Simsekli, Mert Gürbüzbalaban, Sinan Yildirim, Lingjiong Zhu
2024	J	`jnl`	Generalized EXTRA stochastic gradient Langevin dynamics. CoRR Mert Gürbüzbalaban, Mohammad Rafiqul Islam, Xiaoyu Wang, Lingjiong Zhu
2024	J	`jnl`	High Probability and Risk-Averse Guarantees for a Stochastic Accelerated Primal-Dual Method. J. Mach. Learn. Res. Yassine Laguel, Necdet Serhat Aybat, Mert Gürbüzbalaban
2024	A*	`conf`	High-probability complexity bounds for stochastic non-convex minimax optimization. NeurIPS Yassine Laguel, Yasa Syed, Necdet Serhat Aybat, Mert Gürbüzbalaban
2024	J	`jnl`	Penalized Overdamped and Underdamped Langevin Monte Carlo Algorithms for Constrained Sampling. J. Mach. Learn. Res. Mert Gürbüzbalaban, Yuanhan Hu, Lingjiong Zhu
2024	J	`jnl`	Robust Accelerated Primal-Dual Methods for Computing Saddle Points. SIAM J. Optim. Xuan Zhang, Necdet Serhat Aybat, Mert Gürbüzbalaban
2023	J	`jnl`	Accelerated gradient methods for nonconvex optimization: Escape trajectories from strict saddle points and convergence to local minima. CoRR Rishabh Dixit, Mert Gürbüzbalaban, Waheed U. Bajwa
2023	A*	`conf`	Algorithmic Stability of Heavy-Tailed SGD with General Loss Functions. ICML Anant Raj, Lingjiong Zhu, Mert Gürbüzbalaban, Umut Simsekli
2023	J	`jnl`	Algorithmic Stability of Heavy-Tailed SGD with General Loss Functions. CoRR Anant Raj, Lingjiong Zhu, Mert Gürbüzbalaban, Umut Simsekli
2023	B	`conf`	Algorithmic Stability of Heavy-Tailed Stochastic Gradient Descent on Least Squares. ALT Anant Raj, Melih Barsbey, Mert Gürbüzbalaban, Lingjiong Zhu, Umut Simsekli
2023	J	`jnl`	Boundary Conditions for Linear Exit Time Gradient Trajectories Around Saddle Points: Analysis and Algorithm. IEEE Trans. Inf. Theory Rishabh Dixit, Mert Gürbüzbalaban, Waheed U. Bajwa
2023	J	`jnl`	Cyclic and Randomized Stepsizes Invoke Heavier Tails in SGD than Constant Stepsize. Trans. Mach. Learn. Res. Mert Gürbüzbalaban, Yuanhan Hu, Umut Simsekli, Lingjiong Zhu
2023	J	`jnl`	Cyclic and Randomized Stepsizes Invoke Heavier Tails in SGD. CoRR Mert Gürbüzbalaban, Yuanhan Hu, Umut Simsekli, Lingjiong Zhu
2023	A*	`conf`	Uniform-in-Time Wasserstein Stability Bounds for (Noisy) Stochastic Gradient Descent. NeurIPS Lingjiong Zhu, Mert Gürbüzbalaban, Anant Raj, Umut Simsekli
2023	J	`jnl`	Uniform-in-Time Wasserstein Stability Bounds for (Noisy) Stochastic Gradient Descent. CoRR Lingjiong Zhu, Mert Gürbüzbalaban, Anant Raj, Umut Simsekli
2022	J	`jnl`	A Stochastic Subgradient Method for Distributionally Robust Non-convex and Non-smooth Learning. J. Optim. Theory Appl. Mert Gürbüzbalaban, Andrzej Ruszczynski, Landi Zhu
2022	J	`jnl`	A Variance-Reduced Stochastic Accelerated Primal Dual Algorithm. CoRR Bugra Can, Mert Gürbüzbalaban, Necdet Serhat Aybat
2022	J	`jnl`	Algorithmic Stability of Heavy-Tailed Stochastic Gradient Descent on Least Squares. CoRR Anant Raj, Melih Barsbey, Mert Gürbüzbalaban, Lingjiong Zhu, Umut Simsekli
2022	J	`jnl`	Differentially Private Accelerated Optimization Algorithms. SIAM J. Optim. Nurdan Kuru, S. Ilker Birbil, Mert Gürbüzbalaban, Sinan Yildirim
2022	J	`jnl`	Global Convergence of Stochastic Gradient Hamiltonian Monte Carlo for Nonconvex Stochastic Optimization: Nonasymptotic Performance Bounds and Momentum-Based Acceleration. Oper. Res. Xuefeng Gao, Mert Gürbüzbalaban, Lingjiong Zhu
2022	J	`jnl`	Heavy-Tail Phenomenon in Decentralized SGD. CoRR Mert Gürbüzbalaban, Yuanhan Hu, Umut Simsekli, Kun Yuan, Lingjiong Zhu
2022	A	`conf`	HyLo: A Hybrid Low-Rank Natural Gradient Descent Method. SC Baorun Mu, Saeed Soori, Bugra Can, Mert Gürbüzbalaban, Maryam Mehri Dehnavi
2022	J	`jnl`	Penalized Langevin and Hamiltonian Monte Carlo Algorithms for Constrained Sampling. CoRR Mert Gürbüzbalaban, Yuanhan Hu, Lingjiong Zhu
2022	J	`jnl`	Randomized Gossiping With Effective Resistance Weights: Performance Guarantees and Applications. IEEE Trans. Control. Netw. Syst. Bugra Can, Saeed Soori, Necdet Serhat Aybat, Maryam Mehri Dehnavi, Mert Gürbüzbalaban
2022	J	`jnl`	Robust Distributed Accelerated Stochastic Gradient Methods for Multi-Agent Networks. J. Mach. Learn. Res. Alireza Fallah, Mert Gürbüzbalaban, Asuman E. Ozdaglar, Umut Simsekli, Lingjiong Zhu
2022	A*	`conf`	SAPD+: An Accelerated Stochastic Method for Nonconvex-Concave Minimax Problems. NeurIPS Xuan Zhang, Necdet Serhat Aybat, Mert Gürbüzbalaban
2021	A*	`conf`	Asymmetric Heavy Tails and Implicit Bias in Gaussian Noise Injections. ICML Alexander Camuto, Xiaoyu Wang, Lingjiong Zhu, Chris C. Holmes, Mert Gürbüzbalaban, Umut Simsekli
2021	J	`jnl`	Asymmetric Heavy Tails and Implicit Bias in Gaussian Noise Injections. CoRR Alexander Camuto, Xiaoyu Wang, Lingjiong Zhu, Chris C. Holmes, Mert Gürbüzbalaban, Umut Simsekli
2021	A*	`conf`	Convergence Rates of Stochastic Gradient Descent under Infinite Noise Variance. NeurIPS Hongjian Wang, Mert Gürbüzbalaban, Lingjiong Zhu, Umut Simsekli, Murat A. Erdogdu
2021	J	`jnl`	Decentralized Stochastic Gradient Langevin Dynamics and Hamiltonian Monte Carlo. J. Mach. Learn. Res. Mert Gürbüzbalaban, Xuefeng Gao, Yuanhan Hu, Lingjiong Zhu
2021	A*	`conf`	Fractal Structure and Generalization Properties of Stochastic Optimization Algorithms. NeurIPS Alexander Camuto, George Deligiannidis, Murat A. Erdogdu, Mert Gürbüzbalaban, Umut Simsekli, Lingjiong Zhu
2021	J	`jnl`	Fractal Structure and Generalization Properties of Stochastic Optimization Algorithms. CoRR Alexander Camuto, George Deligiannidis, Murat A. Erdogdu, Mert Gürbüzbalaban, Umut Simsekli, Lingjiong Zhu
2021	A	`conf`	Fractional moment-preserving initialization schemes for training deep neural networks. AISTATS Mert Gürbüzbalaban, Yuanhan Hu
2021	—	`conf`	L-DQN: An Asynchronous Limited-Memory Distributed Quasi-Newton Method. CDC Bugra Can, Saeed Soori, Maryam Mehri Dehnavi, Mert Gürbüzbalaban
2021	J	`jnl`	L-DQN: An Asynchronous Limited-Memory Distributed Quasi-Newton Method. CoRR Bugra Can, Saeed Soori, Maryam Mehri Dehnavi, Mert Gürbüzbalaban
2021	J	`jnl`	TENGraD: Time-Efficient Natural Gradient Descent with Exact Fisher-Block Inversion. CoRR Saeed Soori, Bugra Can, Baourun Mu, Mert Gürbüzbalaban, Maryam Mehri Dehnavi
2021	A*	`conf`	The Heavy-Tail Phenomenon in SGD. ICML Mert Gürbüzbalaban, Umut Simsekli, Lingjiong Zhu
2021	J	`jnl`	Why random reshuffling beats stochastic gradient descent. Math. Program. Mert Gürbüzbalaban, Asuman E. Ozdaglar, Pablo A. Parrilo
2020	J	`jnl`	A Stochastic Subgradient Method for Distributionally Robust Non-Convex Learning. CoRR Mert Gürbüzbalaban, Andrzej Ruszczynski, Landi Zhu
2020	A	`conf`	ASYNC: A Cloud Engine with Asynchrony and History for Distributed Machine Learning. IPDPS Saeed Soori, Bugra Can, Mert Gürbüzbalaban, Maryam Mehri Dehnavi
2020	A*	`conf`	Breaking Reversibility Accelerates Langevin Dynamics for Non-Convex Optimization. NeurIPS Xuefeng Gao, Mert Gürbüzbalaban, Lingjiong Zhu
2020	A	`conf`	DAve-QN: A Distributed Averaged Quasi-Newton Method with Local Superlinear Convergence Rate. AISTATS Saeed Soori, Konstantin Mishchenko, Aryan Mokhtari, Maryam Mehri Dehnavi, Mert Gürbüzbalaban
2020	J	`jnl`	Decentralized Stochastic Gradient Langevin Dynamics and Hamiltonian Monte Carlo. CoRR Mert Gürbüzbalaban, Xuefeng Gao, Yuanhan Hu, Lingjiong Zhu
2020	J	`jnl`	Differentially Private Accelerated Optimization Algorithms. CoRR Nurdan Kuru, S. Ilker Birbil, Mert Gürbüzbalaban, Sinan Yildirim
2020	A*	`conf`	Fractional Underdamped Langevin Dynamics: Retargeting SGD with Momentum under Heavy-Tailed Gradient Noise. ICML Umut Simsekli, Lingjiong Zhu, Yee Whye Teh, Mert Gürbüzbalaban
2020	J	`jnl`	Fractional Underdamped Langevin Dynamics: Retargeting SGD with Momentum under Heavy-Tailed Gradient Noise. CoRR Umut Simsekli, Lingjiong Zhu, Yee Whye Teh, Mert Gürbüzbalaban
2020	J	`jnl`	Fractional moment-preserving initialization schemes for training fully-connected neural networks. CoRR Mert Gürbüzbalaban, Yuanhan Hu
2020	A*	`conf`	IDEAL: Inexact DEcentralized Accelerated Augmented Lagrangian Method. NeurIPS Yossi Arjevani, Joan Bruna, Bugra Can, Mert Gürbüzbalaban, Stefanie Jegelka, Hongzhou Lin
2020	J	`jnl`	IDEAL: Inexact DEcentralized Accelerated Augmented Lagrangian Method. CoRR Yossi Arjevani, Joan Bruna, Bugra Can, Mert Gürbüzbalaban, Stefanie Jegelka, Hongzhou Lin
2020	J	`jnl`	Randomness and permutations in coordinate descent methods. Math. Program. Mert Gürbüzbalaban, Asuman E. Ozdaglar, Nuri Denizcan Vanli, Stephen J. Wright
2020	J	`jnl`	Robust Accelerated Gradient Methods for Smooth Strongly Convex Functions. SIAM J. Optim. Necdet Serhat Aybat, Alireza Fallah, Mert Gürbüzbalaban, Asuman E. Ozdaglar
2020	J	`jnl`	The Heavy-Tail Phenomenon in SGD. CoRR Mert Gürbüzbalaban, Umut Simsekli, Lingjiong Zhu
2019	A*	`conf`	A Tail-Index Analysis of Stochastic Gradient Noise in Deep Neural Networks. ICML Umut Simsekli, Levent Sagun, Mert Gürbüzbalaban
2019	J	`jnl`	A Tail-Index Analysis of Stochastic Gradient Noise in Deep Neural Networks. CoRR Umut Simsekli, Levent Sagun, Mert Gürbüzbalaban
2019	A*	`conf`	A Universally Optimal Multistage Accelerated Stochastic Gradient Method. NeurIPS Necdet Serhat Aybat, Alireza Fallah, Mert Gürbüzbalaban, Asuman E. Ozdaglar
2019	J	`jnl`	A Universally Optimal Multistage Accelerated Stochastic Gradient Method. CoRR Necdet Serhat Aybat, Alireza Fallah, Mert Gürbüzbalaban, Asuman E. Ozdaglar
2019	J	`jnl`	ASYNC: Asynchronous Machine Learning on Distributed Systems. CoRR Saeed Soori, Bugra Can, Mert Gürbüzbalaban, Maryam Mehri Dehnavi
2019	A*	`conf`	Accelerated Linear Convergence of Stochastic Momentum Methods in Wasserstein Distances. ICML Bugra Can, Mert Gürbüzbalaban, Lingjiong Zhu
2019	J	`jnl`	Accelerated Linear Convergence of Stochastic Momentum Methods in Wasserstein Distances. CoRR Bugra Can, Mert Gürbüzbalaban, Lingjiong Zhu
2019	J	`jnl`	Convergence Rate of Incremental Gradient and Incremental Newton Methods. SIAM J. Optim. Mert Gürbüzbalaban, Asuman E. Ozdaglar, Pablo A. Parrilo
2019	A*	`conf`	First Exit Time Analysis of Stochastic Gradient Descent Under Heavy-Tailed Gradient Noise. NeurIPS Thanh Huy Nguyen, Umut Simsekli, Mert Gürbüzbalaban, Gaël Richard
2019	J	`jnl`	First Exit Time Analysis of Stochastic Gradient Descent Under Heavy-Tailed Gradient Noise. CoRR Thanh Huy Nguyen, Umut Simsekli, Mert Gürbüzbalaban, Gaël Richard
2019	J	`jnl`	On the Heavy-Tailed Theory of Stochastic Gradient Descent for Deep Neural Networks. CoRR Umut Simsekli, Mert Gürbüzbalaban, Thanh Huy Nguyen, Gaël Richard, Levent Sagun
2019	J	`jnl`	Robust Distributed Accelerated Stochastic Gradient Methods for Multi-Agent Networks. CoRR Alireza Fallah, Mert Gürbüzbalaban, Asuman E. Ozdaglar, Umut Simsekli, Lingjiong Zhu
2018	J	`jnl`	Breaking Reversibility Accelerates Langevin Dynamics for Global Non-Convex Optimization. CoRR Xuefeng Gao, Mert Gürbüzbalaban, Lingjiong Zhu
2018	J	`jnl`	Global Convergence Rate of Proximal Incremental Aggregated Gradient Methods. SIAM J. Optim. N. Denizcan Vanli, Mert Gürbüzbalaban, Asuman E. Ozdaglar
2018	J	`jnl`	Global Convergence of Stochastic Gradient Hamiltonian Monte Carlo for Non-Convex Stochastic Optimization: Non-Asymptotic Performance Bounds and Momentum-Based Acceleration. CoRR Xuefeng Gao, Mert Gürbüzbalaban, Lingjiong Zhu
2018	B	`conf`	Reducing Communication in Proximal Newton Methods for Sparse Least Squares Problems. ICPP Saeed Soori, Aditya Devarakonda, Zachary Blanco, James Demmel, Mert Gürbüzbalaban, Maryam Mehri Dehnavi
2018	J	`jnl`	Robust Accelerated Gradient Methods for Smooth Strongly Convex Functions. CoRR Necdet Serhat Aybat, Alireza Fallah, Mert Gürbüzbalaban, Asuman E. Ozdaglar
2018	J	`jnl`	Surpassing Gradient Descent Provably: A Cyclic Incremental Method with Linear Convergence Rate. SIAM J. Optim. Aryan Mokhtari, Mert Gürbüzbalaban, Alejandro Ribeiro
2017	Misc	`conf`	A double incremental aggregated gradient method with linear convergence rate for large-scale optimization. ICASSP Aryan Mokhtari, Mert Gürbüzbalaban, Alejandro Ribeiro
2017	J	`jnl`	Approximating the Real Structured Stability Radius with Frobenius-Norm Bounded Perturbations. SIAM J. Matrix Anal. Appl. Nicola Guglielmi, Mert Gürbüzbalaban, Tim Mitchell, Michael L. Overton
2017	J	`jnl`	Avoiding Communication in Proximal Methods for Convex Optimization Problems. CoRR Saeed Soori, Aditya Devarakonda, James Demmel, Mert Gürbüzbalaban, Maryam Mehri Dehnavi
2017	—	`conf`	Decentralized computation of effective resistances and acceleration of consensus algorithms. GlobalSIP Necdet Serhat Aybat, Mert Gürbüzbalaban
2017	J	`jnl`	On the Convergence Rate of Incremental Aggregated Gradient Algorithms. SIAM J. Optim. Mert Gürbüzbalaban, Asuman E. Ozdaglar, Pablo A. Parrilo
2017	J	`jnl`	Polynomial root radius optimization with affine constraints. Math. Program. Julia Eaton, Sara Grundel, Mert Gürbüzbalaban, Michael L. Overton
2017	—	`conf`	When Cyclic Coordinate Descent Outperforms Randomized Coordinate Descent. NIPS Mert Gürbüzbalaban, Asuman E. Ozdaglar, Pablo A. Parrilo, Nuri Denizcan Vanli
2016	—	`conf`	Global convergence rate of incremental aggregated gradient methods for nonsmooth problems. CDC N. Denizcan Vanli, Mert Gürbüzbalaban, Asuman E. Ozdaglar
2016	J	`jnl`	Surpassing Gradient Descent Provably: A Cyclic Incremental Method with Linear Convergence Rate. CoRR Aryan Mokhtari, Mert Gürbüzbalaban, Alejandro Ribeiro
2015	J	`jnl`	A globally convergent incremental Newton method. Math. Program. Mert Gürbüzbalaban, Asuman E. Ozdaglar, Pablo A. Parrilo
2013	J	`jnl`	Fast Approximation of the H SIAM J. Matrix Anal. Appl. Nicola Guglielmi, Mert Gürbüzbalaban, Michael L. Overton
2012	J	`jnl`	Explicit Solutions for Root Optimization of a Polynomial Family With One Affine Constraint. IEEE Trans. Autom. Control. Vincent D. Blondel, Mert Gürbüzbalaban, Alexandre Megretski, Michael L. Overton
2012	J	`jnl`	Some Regularity Results for the Pseudospectral Abscissa and Pseudospectral Radius of a Matrix. SIAM J. Optim. Mert Gürbüzbalaban, Michael L. Overton
2010	—	`conf`	Explicit solutions for root optimization of a polynomial family. CDC Vincent D. Blondel, Mert Gürbüzbalaban, Alexandre Megretski, Michael L. Overton