:: Library Catalog

Cover Image

Saved in:

Bibliographic Details
Main Authors:	Zhang, Yikuan, Yang, Ning, Tu, Yuhai
Format:	Preprint
Published:	2026
Subjects:	Machine Learning
Online Access:	https://arxiv.org/abs/2602.05600
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Transient learning dynamics drive escape from sharp valleys in Stochastic Gradient Descent
by: Yang, Ning, et al.
Published: (2026)

Correlating Cross-Iteration Noise for DP-SGD using Model Curvature
by: Gu, Xin, et al.
Published: (2025)

Geometry of optimal control in chemical reaction networks
by: Zhang, Yikuan, et al.
Published: (2024)

Curvature-Aligned Probing for Local Loss-Landscape Stabilization
by: Kiselev, Nikita, et al.
Published: (2026)

A Unified Noise-Curvature View of Loss of Trainability
by: Baveja, Gunbir Singh, et al.
Published: (2025)

The Optimization Landscape of SGD Across the Feature Learning Strength
by: Atanasov, Alexander, et al.
Published: (2024)

Curvature in the Looking-Glass: Optimal Methods to Exploit Curvature of Expectation in the Loss Landscape
by: Duersch, Jed A., et al.
Published: (2024)

Superlinear Multi-Step Attention
by: Huang, Yufeng
Published: (2026)

Curvature-Informed SGD via General Purpose Lie-Group Preconditioners
by: Pooladzandi, Omead, et al.
Published: (2024)

Can SGD Handle Heavy-Tailed Noise?
by: Fatkhullin, Ilyas, et al.
Published: (2025)

SketchySGD: Reliable Stochastic Optimization via Randomized Curvature Estimates
by: Frangella, Zachary, et al.
Published: (2022)

Beyond Implicit Bias: The Insignificance of SGD Noise in Online Learning
by: Vyas, Nikhil, et al.
Published: (2023)

A Scalable Measure of Loss Landscape Curvature for Analyzing the Training Dynamics of LLMs
by: Kalra, Dayal Singh, et al.
Published: (2026)

On the Learning Dynamics of Two-layer Linear Networks with Label Noise SGD
by: Zhang, Tongcheng, et al.
Published: (2026)

Error estimates between SGD with momentum and underdamped Langevin diffusion
by: Guillin, Arnaud, et al.
Published: (2024)

It's Our Loss: No Privacy Amplification for Hidden State DP-SGD With Non-Convex Loss
by: Annamalai, Meenatchi Sundaram Muthu Selva
Published: (2024)

Directions of Curvature as an Explanation for Loss of Plasticity
by: Lewandowski, Alex, et al.
Published: (2023)

Convergent Privacy Loss of Noisy-SGD without Convexity and Smoothness
by: Chien, Eli, et al.
Published: (2024)

Noise is All You Need: Private Second-Order Convergence of Noisy SGD
by: Avdiukhin, Dmitrii, et al.
Published: (2024)

Sensitivity Analysis On Loss Landscape
by: Faroz, Salman
Published: (2024)

Convergence of SGD for Training Neural Networks with Sliced Wasserstein Losses
by: Tanguy, Eloi
Published: (2023)

An Improved Privacy and Utility Analysis of Differentially Private SGD with Bounded Domain and Smooth Losses
by: Liang, Hao, et al.
Published: (2025)

A Precise Characterization of SGD Stability Using Loss Surface Geometry
by: Dexter, Gregory, et al.
Published: (2024)

Global Convergence of SGD For Logistic Loss on Two Layer Neural Nets
by: Gopalani, Pulkit, et al.
Published: (2023)

Expectation Consistency Loss: Rethink Confidence Calibration under Covariate Shift
by: Dong, Jinzong, et al.
Published: (2026)

Population Risk Bounds for Kolmogorov-Arnold Networks Trained by DP-SGD with Correlated Noise
by: Wang, Puyu, et al.
Published: (2026)

Adam or Gauss-Newton? A Comparative Study In Terms of Basis Alignment and SGD Noise
by: Liu, Bingbin, et al.
Published: (2025)

Curvature Clues: Decoding Deep Learning Privacy with Input Loss Curvature
by: Ravikumar, Deepak, et al.
Published: (2024)

Incremental Gauss--Newton Methods with Superlinear Convergence Rates
by: Zhou, Zhiling, et al.
Published: (2024)

LossLens: Diagnostics for Machine Learning through Loss Landscape Visual Analytics
by: Xie, Tiankai, et al.
Published: (2024)

Landscaper: Understanding Loss Landscapes Through Multi-Dimensional Topological Analysis
by: Chen, Jiaqing, et al.
Published: (2026)

There is a Singularity in the Loss Landscape
by: Lowell, Mark
Published: (2022)

Exact Risk Curves of signSGD in High-Dimensions: Quantifying Preconditioning and Noise-Compression Effects
by: Xiao, Ke Liang, et al.
Published: (2024)

Hardware-Adaptive and Superlinear-Capacity Memristor-based Associative Memory
by: He, Chengping, et al.
Published: (2025)

Incremental Quasi-Newton Methods with Faster Superlinear Convergence Rates
by: Liu, Zhuanghua, et al.
Published: (2024)

Visualizing Loss Functions as Topological Landscape Profiles
by: Geniesse, Caleb, et al.
Published: (2024)

From Memorization to Reasoning in the Spectrum of Loss Curvature
by: Merullo, Jack, et al.
Published: (2025)

Paths and Ambient Spaces in Neural Loss Landscapes
by: Dold, Daniel, et al.
Published: (2025)

Using Degeneracy in the Loss Landscape for Mechanistic Interpretability
by: Bushnaq, Lucius, et al.
Published: (2024)

Stochastic Collapse: How Gradient Noise Attracts SGD Dynamics Towards Simpler Subnetworks
by: Chen, Feng, et al.
Published: (2023)