NeurIPS 2026 reviews are up: reviews, scores, rebuttals and meta-reviews for 7,877 papers. See the highest rated
acceptodds: the papers of ICLR 2027
98 results for “keyword:"gradient descent"” in ICLR 2027. 9 withdrawn or desk rejected listed apart. clearsort:relevanceratingnewestmost read
[PDF]What Drives the Inlier-Memorization Effect? A Theory of Outlier Detection via Early Training Dynamics▲9.2k2.67
[PDF]Implicit Sparse Regularization in Momentum Gradient Descent: A Shifted Early Stopping Window Induced by Momentum and Depth▼4.3k2.00
[PDF]Convergence Guarantees of Gradient Descent for Neural Networks via Generalized Lipschitz Smoothness▲3972.67
[PDF]Stability-Based Adversarial Generalization Bounds for Hypergraph Convolutional Networks▼4.3k2.00
[PDF]Optimal Recursive Composition and Dyadic Phase Laws for Gradient Descent with Predetermined Stepsizes▲5103.00
[PDF]Bringing Order to Asynchronous SGD: Towards Optimality under Data-Dependent Delays with Momentum▼5.9k2.50
[PDF]Checkable NTK Positivity and Finite-Width Gradient Descent for Scalar- and Vector-Valued PINNs with Strong-Form, Weak-Form, and Nonlocal Linear Constraints▼4.3k2.75
[PDF]The Exact Time-Uniform Rate Frontier for Stochastic Gradient Descent on Smooth Convex Objectives▼4.5k2.67
[PDF]Implicit Low-Rank Training Dynamics in Directly Parameterized Lipschitz-Bounded deep Networks▲5103.00
[PDF]SGD Selects Initialization-Dependent Representation in Overparameterized ReLU Networks▼4.5k2.67
[PDF]Momentum Makes (Adaptive) Subgradient Methods Optimal for Stochastic Nonsmooth Nonconvex Optimization▼5.8k2.33
[PDF]Rank-1 Factored Gradient Descent without Restricted Isometry: Application to Quantum Pure State Tomography on Multi-Qubits▼4.3k2.00
[PDF]From Conservative Certificates to Adaptive Capital: Faster Anytime-Valid Stopping for SGD▲3.0k2.33
[PDF]On The Rate Of Convergence Of GD In Non Linear Neural Networks: An Adversarial Robustness Perspective▼1.2k1.67
[PDF]Provable Benefits of Block Preconditioning over Gradient Descent under Block-Heterogeneous Smoothness▼4.5k2.67
Withdrawn or desk rejected (9)
[PDF]Training-Loss Guarantees for Muon with Finite-Step Newton-Schulz OrthogonalizationWithdrawn1.67
[PDF]SGD-Induced Parameter Drift in the Saddle-to-Saddle Dynamics of Deep Linear NetworksWithdrawn2.33