Skip to content
conftrace_

Itay Safran

11 papers · 2016–2026 · 3 venues · across top CS/AI venues

Authors are matched by name, so researchers who share a name ("Yang Liu", "Wei Wang") appear as a single profile and their paper counts, venues and achievements are combined. How identity is handled.

Achievements

Jump to papers ↓
+6 more ↓ πŸ—ΊοΈ Taxonomy Completionist (12) πŸƒ Academic Marathon Β· Half 🌍 Conference Polyglot Β· Bronze πŸŒ‰ Interdisciplinary Bridge Β· Bronze 🐣 Hot Topic Early Bird
🧬 Topic Evolution 🧭 Keyword Pioneer πŸ—ƒοΈ Keyword Collector (58) πŸ’Ž Century Club Β· Bronze πŸ”₯ Unstoppable Β· Bronze ❓ The Questioner (2)

Venues

COLT (6) ICML (3) NeurIPS (2)

Top co-authors

Ohad Shamir (6) Paul Valiant (2) Jason Lee (2) Abhigyan Dutta (1) Daniel Reichman (1) Gal Vardi (1) Ronen Eldan (1)

Research topics

Deep Learning (10) Architectures (9) Machine Learning (8) Optimization & Theory Β· Machine Learning (8) Optimization & Theory Β· Deep Learning (7) Mathematics & Optimization (3) Optimization (2) Computer Science (1) Foundations (1) Mathematics (1)

Keywords

relu network (3) depth separation (3) function approximation (3) convex optimization (2) stochastic gradient descent (2) neural network (2) depth-width tradeoff (2) convergence analysis (1) gradient descent (1) finite-sum optimization (1) loss landscape (1) strongly convex (1) approximation error (1) optimization theory (1) gradient flow (1) initialization (1) convergence rate (1) convergence guarantee (1) overspecified networks (1) generalization bound (1)

Papers

A Depth Hierarchy for Computing the Maximum in ReLU Networks via Extremal Graph Theory COLT 2026 The Median is Easier than it Looks: Approximation with a Constant-Depth, Linear-Width ReLU Network COLT 2026 Depth Separations in Neural Networks: Separating the Dimension from the Accuracy COLT 2025 On the Effective Number of Linear Regions in Shallow Univariate ReLU Networks: Convergence Guarantees and Implicit Bias NeurIPS 2022 Optimization-Based Separations for Neural Networks COLT 2022 Random Shuffling Beats SGD Only After Many Epochs on Ill-Conditioned Problems NeurIPS 2021 How Good is SGD with Random Shuffling? COLT 2020 Depth Separations in Neural Networks: What is Actually Being Separated? COLT 2019 Spurious Local Minima are Common in Two-Layer ReLU Neural Networks ICML 2018 Depth-Width Tradeoffs in Approximating Natural Functions with Neural Networks ICML 2017 On the Quality of the Initial Basin in Overspecified Neural Networks ICML 2016