NS
Nathan Srebro
cs.LGstat.MLmath.OCcs.AImath.STstat.THcs.DScs.CCcs.CLcs.CV
On Valency
published · living versionsW_rfzhcwyd·v1 · currentpublished
Global Convergence in Deep Networks via the Second Law of Thermodynamics
with Matan Tsipory, Ofir Gaash, Itamar Harel, Daniel Soudry
1 version
Preprints & journals
157 papers in the corpus · 2003–2026Global Convergence in Deep Networks via the Second Law of Thermodynamics2610.24775v1 · Matan Tsipory, Ofir Gaash, Itamar Harel et al.2026 · 0 citationsarXivon Valency
Recursive Language Models Generalize Out of Domain2609.20831v1 · Chenxiao Yang, Zhiyuan Li, David McAllester et al.2026 · 0 citationsarXiv
Hierarchical Domain Generalization2607.16528v1 · Chenxiao Yang, Zhiyuan Li, Shai Ben-David et al.2026 · 0 citationsarXiv
On Incentivized Exploration beyond Bayesianism and Full-Information2607.18300v1 · Dimitar Chakarov, Lee Cohen, Nathan Srebro2026 · 0 citationsarXiv
Tight Sample Complexity of Transformers2606.09731v2 · Chenxiao Yang, Nathan Srebro, Zhiyuan Li2026 · 0 citationsarXiv
Learning to Think from Multiple Thinkers2604.24737v1 · Nirmit Joshi, Roey Magen, Nathan Srebro et al.2026 · 0 citationsarXiv
Overfitting and Generalizing with (PAC) Bayesian Prediction in Noisy Binary Classification2603.22644v1 · Xiaohan Zhu, Mesrob I. Ohannessian, Nathan Srebro2026 · 0 citationsarXiv
Recursive Models for Long-Horizon Reasoning2603.02112v2 · Chenxiao Yang, Nathan Srebro, Zhiyuan Li2026 · 0 citationsarXiv
Positive Distribution Shift as a Framework for Understanding Tractable Learning2602.08907v2 · Marko Medvedev, Idan Attias, Elisabetta Cornacchia et al.2026 · 0 citationsarXiv
Research Program: Theory of Learning in Dynamical Systems2512.19410v1 · Elad Hazan, Shai Shalev Shwartz, Nathan Srebro2025 · 0 citationsarXiv
Shift is Good: Mismatched Data Mixing Improves Test Performance2510.25108v2 · Marko Medvedev, Kaifeng Lyu, Zhiyuan Li et al.2025 · 0 citationsarXiv
Learning to Answer from Correct Demonstrations2510.15464v2 · Nirmit Joshi, Gene Li, Siddharth Bhandari et al.2025 · 0 citationsarXiv
On the Hardness of Learning Regular Expressions2510.04834v1 · Idan Attias, Lev Reyzin, Nathan Srebro et al.2025 · 0 citationsarXiv
Learning single-index models via harmonic decomposition2506.09887v2 · Nirmit Joshi, Hugo Koubbi, Theodor Misiakiewicz et al.2025 · 0 citationsarXiv
Temperature is All You Need for Generalization in Langevin Dynamics and other Markov Processes2505.19087v2 · Itamar Harel, Yonathan Wolanowsky, Gal Vardi et al.2025 · 0 citationsarXiv
From Continual Learning to SGD and Back: Better Rates for Continual Linear Models2504.04579v3 · Itay Evron, Ran Levinstein, Matan Schliserman et al.2025 · 0 citationsarXiv
PENCIL: Long Thoughts with Short Memory2503.14337v2 · Chenxiao Yang, Nathan Srebro, David McAllester et al.2025 · 0 citationsarXiv
A Theory of Learning with Autoregressive Chain of Thought2503.07932v2 · Nirmit Joshi, Gal Vardi, Adam Block et al.2025 · 0 citationsarXiv
Weak-to-Strong Generalization Even in Random Feature Networks, Provably2503.02877v3 · Marko Medvedev, Kaifeng Lyu, Dingli Yu et al.2025 · 0 citationsarXiv
Quantifying Overfitting along the Regularization Path for Two-Part-Code MDL in Supervised Classification2503.02110v2 · Xiaohan Zhu, Nathan Srebro2025 · 0 citationsarXiv
Tight Bounds on the Binomial CDF, and the Minimum of i.i.d Binomials, in terms of KL-Divergence2502.18611v1 · Xiaohan Zhu, Mesrob I. Ohannessian, Nathan Srebro2025 · 0 citationsarXiv
Provable Tempered Overfitting of Minimal Nets and Typical Nets2410.19092v1 · Itamar Harel, William M. Hoza, Gal Vardi et al.2024 · 1 citationarXiv
Score Design for Multi-Criteria Incentivization2410.06290v1 · Anmol Kabra, Mina Karzand, Tosca Lechner et al.2024 · 0 citationsarXiv
Overfitting Behaviour of Gaussian Kernel Ridgeless Regression: Varying Bandwidth or Dimensionality2409.03891v1 · Marko Medvedev, Gal Vardi, Nathan Srebro2024 · 2 citationsarXiv
On the Complexity of Learning Sparse Functions with Statistical and Gradient Queries2407.05622v1 · Nirmit Joshi, Theodor Misiakiewicz, Nathan Srebro2024 · 1 citationarXiv
The Price of Implicit Bias in Adversarially Robust Generalization2406.04981v1 · Nikolaos Tsilivis, Natalie Frank, Nathan Srebro et al.2024 · 0 citationsarXiv
The Limits and Potentials of Local SGD for Distributed Heterogeneous Learning with Intermittent Communication2405.11667v2 · Kumar Kshitij Patel, Margalit Glasgow, Ali Zindari et al.2024 · 2 citationsarXiv
Depth Separation in Norm-Bounded Infinite-Width Neural Networks2402.08808v1 · Suzanna Parkinson, Greg Ongie, Rebecca Willett et al.2024 · 0 citationsarXiv
How Uniform Random Weights Induce Non-uniform Bias: Typical Interpolating Neural Networks Generalize with Narrow Teachers2402.06323v3 · Gon Buzaglo, Itamar Harel, Mor Shpigel Nacson et al.2024 · 0 citationsarXiv
Metalearning with Very Few Samples Per Task2312.13978v2 · Maryam Aliakbarpour, Konstantina Bairaktari, Gavin Brown et al.2023 · 0 citationsarXiv
Applying statistical learning theory to deep learning2311.15404v2 · C'edric Gerbelot, Avetik Karagulyan, Stefani Karp et al.2023 · 6 citationsarXiv
When is Agnostic Reinforcement Learning Statistically Tractable?2310.06113v1 · Zeyu Jia, Gene Li, Alexander Rakhlin et al.2023 · 0 citationsarXiv
Noisy Interpolation Learning with Shallow Univariate ReLU Networks2307.15396v3 · Nirmit Joshi, Gal Vardi, Nathan Srebro2023 · 0 citationsarXiv
An Agnostic View on the Cost of Overfitting in (Kernel) Ridge Regression2306.13185v2 · Lijia Zhou, James B. Simon, Gal Vardi et al.2023 · 1 citationarXiv
Uniform Convergence with Square-Root Lipschitz Loss2306.13188v1 · Lijia Zhou, Zhen Dai, Frederic Koehler et al.2023 · 0 citationsarXiv
Continual Learning in Linear Classification on Separable Data2306.03534v1 · Itay Evron, Edward Moroshko, Gon Buzaglo et al.2023 · 3 citationsarXiv
Most Neural Networks Are Almost Learnable2305.16508v3 · Amit Daniely, Nathan Srebro, Gal Vardi2023 · 0 citationsarXiv
The Double-Edged Sword of Implicit Bias: Generalization vs. Robustness in ReLU Networks2303.01456v2 · Spencer Frei, Gal Vardi, Peter L. Bartlett et al.2023 · 2 citationsarXiv
Benign Overfitting in Linear Classifiers and Leaky ReLU Networks from KKT Conditions for Margin Maximization2303.01462v1 · Spencer Frei, Gal Vardi, Peter L. Bartlett et al.2023 · 3 citationsarXiv
Computational Complexity of Learning Neural Networks: Smoothness and Degeneracy2302.07426v2 · Amit Daniely, Nathan Srebro, Gal Vardi2023 · 2 citationsarXiv
Interpolation Learning With Minimum Description Length2302.07263v1 · Naren Sarayu Manoj, Nathan Srebro2023 · 0 citationsarXiv
A Non-Asymptotic Moreau Envelope Theory for High-Dimensional Generalized Linear Models2210.12082v1 · Lijia Zhou, Frederic Koehler, Pragya Sur et al.2022 · 3 citationsarXiv
Implicit Bias in Leaky ReLU Networks Trained on High-Dimensional Data2210.07082v1 · Spencer Frei, Gal Vardi, Peter L. Bartlett et al.2022 · 5 citationsarXiv
Adversarially Robust Learning: A Generic Minimax Optimal Learner and Characterization2209.07369v1 · Omar Montasser, Steve Hanneke, Nathan Srebro2022 · 0 citationsarXiv
Pessimism for Offline Linear Contextual Bandits using Confidence Sets2205.10671v2 · Gene Li, Cong Ma, Nathan Srebro2022 · 0 citationsarXiv
Thinking Outside the Ball: Optimal Learning with Gradient Descent for Generalized Linear Stochastic Convex Optimization2202.13328v2 · Idan Amir, Roi Livni, Nathan Srebro2022 · 0 citationsarXiv
The Sample Complexity of One-Hidden-Layer Neural Networks2202.06233v2 · Gal Vardi, Ohad Shamir, Nathan Srebro2022 · 2 citationsarXiv
Exponential Family Model-Based Reinforcement Learning via Score Matching2112.14195v2 · Gene Li, Junbo Li, Anmol Kabra et al.2021 · 3 citationsarXiv
Optimistic Rates: A Unifying Theory for Interpolation Learning and Regularization in Linear Regression2112.04470v1 · Lijia Zhou, Frederic Koehler, Danica J. Sutherland et al.2021 · 4 citationsarXiv
Transductive Robust Learning Guarantees2110.10602v1 · Omar Montasser, Steve Hanneke, Nathan Srebro2021 · 1 citationarXiv
Career total: 297 works. 157 are in this corpus.Showing the 50 most recent.
Profile built from the corpus for this byline.
Author records are still filling in while the Hub is in alpha. If this is your page, you'll be able to claim it soon. Spot a mistake? Tell us.