Murat A. Erdogdu

86 papers A* 35A 2B 1Misc 1Journal 40Unranked 7

Year	Rank	Type	Title / Venue / Authors
2026	J	`jnl`	Post-Training with Policy Gradients: Optimality and the Base Model Barrier. CoRR Alireza Mousavi Hosseini, Murat A. Erdogdu
2025	J	`jnl`	Analysis of Langevin Monte Carlo from Poincaré to Log-Sobolev. Found. Comput. Math. Sinho Chewi, Murat A. Erdogdu, Mufan (Bill) Li, Ruoqi Shen, Matthew Shunshi Zhang
2025	A*	`conf`	Categorical Distributional Reinforcement Learning with Kullback-Leibler Divergence: Convergence and Asymptotics. ICML Tyler Kastner, Mark Rowland, Yunhao Tang, Murat A. Erdogdu, Amir-massoud Farahmand
2025	J	`jnl`	Distributional Training Data Attribution. CoRR Bruno Mlodozeniec, Isaac Reid, Sam Power, David Krueger, Murat A. Erdogdu, Richard E. Turner, Roger B. Grosse
2025	J	`jnl`	From Information to Generative Exponent: Learning Rate Induces Phase Transitions in SGD. CoRR Konstantinos Christopher Tsiolis, Alireza Mousavi Hosseini, Murat A. Erdogdu
2025	A*	`conf`	Learning Multi-Index Models with Neural Networks via Mean-Field Langevin Dynamics. ICLR Alireza Mousavi Hosseini, Denny Wu, Murat A. Erdogdu
2025	A*	`conf`	Robust Feature Learning for Multi-Index Models in High Dimensions. ICLR Alireza Mousavi Hosseini, Adel Javanmard, Murat A. Erdogdu
2025	J	`jnl`	When Do Transformers Outperform Feedforward and Recurrent Networks? A Statistical Perspective. CoRR Alireza Mousavi Hosseini, Clayton Sanford, Denny Wu, Murat A. Erdogdu
2024	A*	`conf`	A Separation in Heavy-Tailed Sampling: Gaussian vs. Stable Oracles for Proximal Samplers. NeurIPS Ye He, Alireza Mousavi Hosseini, Krishnakumar Balasubramanian, Murat A. Erdogdu
2024	J	`jnl`	An Analysis of Transformed Unadjusted Langevin Algorithm for Heavy-Tailed Sampling. IEEE Trans. Inf. Theory Ye He, Krishnakumar Balasubramanian, Murat A. Erdogdu
2024	J	`jnl`	Beyond Labeling Oracles - What does it mean to steal ML models? Trans. Mach. Learn. Res. Avital Shafran, Ilia Shumailov, Murat A. Erdogdu, Nicolas Papernot
2024	J	`jnl`	Learning Multi-Index Models with Neural Networks via Mean-Field Langevin Dynamics. CoRR Alireza Mousavi Hosseini, Denny Wu, Murat A. Erdogdu
2024	J	`jnl`	Mean-Square Analysis of Discretized Itô Diffusions for Heavy-tailed Sampling. J. Mach. Learn. Res. Ye He, Tyler Farghly, Krishnakumar Balasubramanian, Murat A. Erdogdu
2024	A*	`conf`	Minimax Linear Regression under the Quantile Risk. COLT Ayoub El Hanchi, Chris J. Maddison, Murat A. Erdogdu
2024	A*	`conf`	On the Efficiency of ERM in Feature Learning. NeurIPS Ayoub El Hanchi, Chris J. Maddison, Murat A. Erdogdu
2024	J	`jnl`	On the Efficiency of ERM in Feature Learning. CoRR Ayoub El Hanchi, Chris J. Maddison, Murat A. Erdogdu
2024	A*	`conf`	Pruning is Optimal for Learning Sparse Features in High-Dimensions. COLT Nuri Mert Vural, Murat A. Erdogdu
2024	J	`jnl`	Pruning is Optimal for Learning Sparse Features in High-Dimensions. CoRR Nuri Mert Vural, Murat A. Erdogdu
2024	J	`jnl`	Robust Feature Learning for Multi-Index Models in High Dimensions. CoRR Alireza Mousavi Hosseini, Adel Javanmard, Murat A. Erdogdu
2024	A*	`conf`	Sampling from the Mean-Field Stationary Distribution. COLT Yunbum Kook, Matthew Shunshi Zhang, Sinho Chewi, Murat A. Erdogdu, Mufan (Bill) Li
2024	J	`jnl`	Sampling from the Mean-Field Stationary Distribution. CoRR Yunbum Kook, Matthew Shunshi Zhang, Sinho Chewi, Murat A. Erdogdu, Mufan Bill Li
2023	J	`jnl`	Beyond Labeling Oracles: What does it mean to steal ML models? CoRR Avital Shafran, Ilia Shumailov, Murat A. Erdogdu, Nicolas Papernot
2023	A*	`conf`	Distributional Model Equivalence for Risk-Sensitive Reinforcement Learning. NeurIPS Tyler Kastner, Murat A. Erdogdu, Amir-massoud Farahmand
2023	J	`jnl`	Distributional Model Equivalence for Risk-Sensitive Reinforcement Learning. CoRR Tyler Kastner, Murat A. Erdogdu, Amir-massoud Farahmand
2023	A*	`conf`	Gradient-Based Feature Learning under Structured Data. NeurIPS Alireza Mousavi Hosseini, Denny Wu, Taiji Suzuki, Murat A. Erdogdu
2023	J	`jnl`	Gradient-Based Feature Learning under Structured Data. CoRR Alireza Mousavi Hosseini, Denny Wu, Taiji Suzuki, Murat A. Erdogdu
2023	A*	`conf`	Improved Discretization Analysis for Underdamped Langevin Monte Carlo. COLT Matthew Shunshi Zhang, Sinho Chewi, Mufan (Bill) Li, Krishna Balasubramanian, Murat A. Erdogdu
2023	A*	`conf`	Learning in the Presence of Low-dimensional Structure: A Spiked Random Matrix Perspective. NeurIPS Jimmy Ba, Murat A. Erdogdu, Taiji Suzuki, Zhichao Wang, Denny Wu
2023	J	`jnl`	Mean-Square Analysis of Discretized Itô Diffusions for Heavy-tailed Sampling. CoRR Ye He, Tyler Farghly, Krishnakumar Balasubramanian, Murat A. Erdogdu
2023	A*	`conf`	Neural Networks Efficiently Learn Low-Dimensional Representations with SGD. ICLR Alireza Mousavi Hosseini, Sejun Park, Manuela Girotti, Ioannis Mitliagkas, Murat A. Erdogdu
2023	A*	`conf`	Optimal Excess Risk Bounds for Empirical Risk Minimization on p-Norm Linear Regression. NeurIPS Ayoub El Hanchi, Murat A. Erdogdu
2023	A*	`conf`	Towards a Complete Analysis of Langevin Monte Carlo: Beyond Poincaré Inequality. COLT Alireza Mousavi Hosseini, Tyler K. Farghly, Ye He, Krishna Balasubramanian, Murat A. Erdogdu
2022	A*	`conf`	Analysis of Langevin Monte Carlo from Poincare to Log-Sobolev. COLT Sinho Chewi, Murat A. Erdogdu, Mufan (Bill) Li, Ruoqi Shen, Shunshi Zhang
2022	A*	`conf`	Convergence and Optimality of Policy Gradient Methods in Weakly Smooth Settings. AAAI Matthew Shunshi Zhang, Murat A. Erdogdu, Animesh Garg
2022	A	`conf`	Convergence of Langevin Monte Carlo in Chi-Squared and Rényi Divergence. AISTATS Murat A. Erdogdu, Rasa Hosseinzadeh, Shunshi Zhang
2022	J	`jnl`	Convergence rate of block-coordinate maximization Burer-Monteiro method for solving large SDPs. Math. Program. Murat A. Erdogdu, Asuman E. Ozdaglar, Pablo A. Parrilo, Nuri Denizcan Vanli
2022	A*	`conf`	Generalization Bounds for Stochastic Gradient Descent via Localized $\varepsilon$-Covers. NeurIPS Sejun Park, Umut Simsekli, Murat A. Erdogdu
2022	J	`jnl`	Generalization Bounds for Stochastic Gradient Descent via Localized ε-Covers. CoRR Sejun Park, Umut Simsekli, Murat A. Erdogdu
2022	A*	`conf`	High-dimensional Asymptotics of Feature Learning: How One Gradient Step Improves the Representation. NeurIPS Jimmy Ba, Murat A. Erdogdu, Taiji Suzuki, Zhichao Wang, Denny Wu, Greg Yang
2022	J	`jnl`	High-dimensional Asymptotics of Feature Learning: How One Gradient Step Improves the Representation. CoRR Jimmy Ba, Murat A. Erdogdu, Taiji Suzuki, Zhichao Wang, Denny Wu, Greg Yang
2022	A*	`conf`	Mirror Descent Strikes Again: Optimal Stochastic Convex Optimization under Infinite Noise Variance. COLT Nuri Mert Vural, Lu Yu, Krishnakumar Balasubramanian, Stanislav Volgushev, Murat A. Erdogdu
2022	J	`jnl`	Mirror Descent Strikes Again: Optimal Stochastic Convex Optimization under Infinite Noise Variance. CoRR Nuri Mert Vural, Lu Yu, Krishnakumar Balasubramanian, Stanislav Volgushev, Murat A. Erdogdu
2022	J	`jnl`	Neural Networks Efficiently Learn Low-Dimensional Representations with SGD. CoRR Alireza Mousavi Hosseini, Sejun Park, Manuela Girotti, Ioannis Mitliagkas, Murat A. Erdogdu
2022	A*	`conf`	Towards a Theory of Non-Log-Concave Sampling: First-Order Stationarity Guarantees for Langevin Monte Carlo. COLT Krishna Balasubramanian, Sinho Chewi, Murat A. Erdogdu, Adil Salim, Shunshi Zhang
2022	A*	`conf`	Understanding the Variance Collapse of SVGD in High Dimensions. ICLR Jimmy Ba, Murat A. Erdogdu, Marzyeh Ghassemi, Shengyang Sun, Taiji Suzuki, Denny Wu, Tianzong Zhang
2022	J	`jnl`	p-DkNN: Out-of-Distribution Detection Through Statistical Testing of Deep Representations. CoRR Adam Dziedzic, Stephan Rabanser, Mohammad Yaghini, Armin Ale, Murat A. Erdogdu, Nicolas Papernot
2021	A*	`conf`	An Analysis of Constant Step Size SGD in the Non-convex Regime: Asymptotic Normality and Bias. NeurIPS Lu Yu, Krishnakumar Balasubramanian, Stanislav Volgushev, Murat A. Erdogdu
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`	Convergence and Optimality of Policy Gradient Methods in Weakly Smooth Settings. CoRR Matthew Shunshi Zhang, Murat A. Erdogdu, Animesh Garg
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`	Heavy Tails in SGD and Compressibility of Overparametrized Neural Networks. NeurIPS Melih Barsbey, Milad Sefidgaran, Murat A. Erdogdu, Gaël Richard, Umut Simsekli
2021	J	`jnl`	Heavy Tails in SGD and Compressibility of Overparametrized Neural Networks. CoRR Melih Barsbey, Milad Sefidgaran, Murat A. Erdogdu, Gaël Richard, Umut Simsekli
2021	A*	`conf`	Manipulating SGD with Data Ordering Attacks. NeurIPS Ilia Shumailov, Zakhar Shumaylov, Dmitry Kazhdan, Yiren Zhao, Nicolas Papernot, Murat A. Erdogdu, Ross J. Anderson
2021	J	`jnl`	Manipulating SGD with Data Ordering Attacks. CoRR Ilia Shumailov, Zakhar Shumaylov, Dmitry Kazhdan, Yiren Zhao, Nicolas Papernot, Murat A. Erdogdu, Ross J. Anderson
2021	A*	`conf`	On Empirical Risk Minimization with Dependent and Heavy-Tailed Data. NeurIPS Abhishek Roy, Krishnakumar Balasubramanian, Murat A. Erdogdu
2021	A*	`conf`	On the Convergence of Langevin Monte Carlo: The Interplay between Tail Growth and Smoothness. COLT Murat A. Erdogdu, Rasa Hosseinzadeh
2020	J	`jnl`	A Brief Note on the Convergence of Langevin Monte Carlo in Chi-Square Divergence. CoRR Murat A. Erdogdu, Rasa Hosseinzadeh
2020	J	`jnl`	An Analysis of Constant Step Size SGD in the Non-convex Regime: Asymptotic Normality and Bias. CoRR Lu Yu, Krishnakumar Balasubramanian, Stanislav Volgushev, Murat A. Erdogdu
2020	A*	`conf`	Generalization of Two-layer Neural Networks: An Asymptotic Viewpoint. ICLR Jimmy Ba, Murat A. Erdogdu, Taiji Suzuki, Denny Wu, Tianzong Zhang
2020	A*	`conf`	Hausdorff Dimension, Heavy Tails, and Generalization in Neural Networks. NeurIPS Umut Simsekli, Ozan Sener, George Deligiannidis, Murat A. Erdogdu
2020	J	`jnl`	Hausdorff Dimension, Stochastic Differential Equations, and Generalization in Neural Networks. CoRR Umut Simsekli, Ozan Sener, George Deligiannidis, Murat A. Erdogdu
2020	J	`jnl`	On the Convergence of Langevin Monte Carlo: The Interplay between Tail Growth and Smoothness. CoRR Murat A. Erdogdu, Rasa Hosseinzadeh
2020	A*	`conf`	On the Ergodicity, Bias and Asymptotic Normality of Randomized Midpoint Sampling Method. NeurIPS Ye He, Krishnakumar Balasubramanian, Murat A. Erdogdu
2020	J	`jnl`	On the Ergodicity, Bias and Asymptotic Normality of Randomized Midpoint Sampling Method. CoRR Ye He, Krishnakumar Balasubramanian, Murat A. Erdogdu
2020	J	`jnl`	Riemannian Langevin Algorithm for Solving Semidefinite Programs. CoRR Mufan (Bill) Li, Murat A. Erdogdu
2019	A*	`conf`	Normal Approximation for Stochastic Gradient Descent via Non-Asymptotic Rates of Martingale CLT. COLT Andreas Anastasiou, Krishnakumar Balasubramanian, Murat A. Erdogdu
2019	J	`jnl`	Scalable Approximations for Generalized Linear Problems. J. Mach. Learn. Res. Murat A. Erdogdu, Mohsen Bayati, Lee H. Dicker
2019	J	`jnl`	Stochastic Runge-Kutta Accelerates Langevin Monte Carlo and Beyond. CoRR Xuechen Li, Denny Wu, Lester Mackey, Murat A. Erdogdu
2018	J	`jnl`	Convergence Rate of Block-Coordinate Maximization Burer-Monteiro Method for Solving Large SDPs. CoRR Murat A. Erdogdu, Asuman E. Ozdaglar, Pablo A. Parrilo, Nuri Denizcan Vanli
2018	A*	`conf`	Global Non-convex Optimization with Discretized Diffusions. NeurIPS Murat A. Erdogdu, Lester Mackey, Ohad Shamir
2018	J	`jnl`	Global Non-convex Optimization with Discretized Diffusions. CoRR Murat A. Erdogdu, Lester Mackey, Ohad Shamir
2017	Misc	`conf`	Generalized Hessian approximations via Stein's lemma for constrained minimization. ITA Murat A. Erdogdu
2017	—	`conf`	Inference in Graphical Models via Semidefinite Programming Hierarchies. NIPS Murat A. Erdogdu, Yash Deshpande, Andrea Montanari
2017	J	`jnl`	Inference in Graphical Models via Semidefinite Programming Hierarchies. CoRR Murat A. Erdogdu, Yash Deshpande, Andrea Montanari
2017	—	`conf`	Robust Estimation of Neural Signals in Calcium Imaging. NIPS Hakan Inan, Murat A. Erdogdu, Mark J. Schnitzer
2016	A	`conf`	Maximum Likelihood for Variance Estimation in High-Dimensional Linear Models. AISTATS Lee H. Dicker, Murat A. Erdogdu
2016	J	`jnl`	Newton-Stein Method: An Optimization Method for GLMs via Stein's Lemma. J. Mach. Learn. Res. Murat A. Erdogdu
2016	—	`conf`	Scaled Least Squares Estimator for GLMs in Large-Scale Problems. NIPS Murat A. Erdogdu, Lee H. Dicker, Mohsen Bayati
2015	—	`conf`	Convergence rates of sub-sampled Newton methods. NIPS Murat A. Erdogdu, Andrea Montanari
2015	—	`conf`	Newton-Stein Method: A Second Order Method for GLMs via Stein's Lemma. NIPS Murat A. Erdogdu
2015	—	`conf`	Privacy-Utility Trade-Off for Time-Series with Application to Smart-Meter Data. AAAI Workshop: Computational Sustainability Murat A. Erdogdu, Nadia Fawaz, Andrea Montanari
2015	B	`conf`	Privacy-utility trade-off under continual observation. ISIT Murat A. Erdogdu, Nadia Fawaz
2015	A*	`conf`	SEISMIC: A Self-Exciting Point Process Model for Predicting Tweet Popularity. KDD Qingyuan Zhao, Murat A. Erdogdu, Hera Y. He, Anand Rajaraman, Jure Leskovec
2015	J	`jnl`	SEISMIC: A Self-Exciting Point Process Model for Predicting Tweet Popularity. CoRR Qingyuan Zhao, Murat A. Erdogdu, Hera Y. He, Anand Rajaraman, Jure Leskovec
2013	—	`conf`	Estimating LASSO Risk and Noise Level. NIPS Mohsen Bayati, Murat A. Erdogdu, Andrea Montanari