Notice: The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty and potential bias; scores should be interpreted as estimates. Full accuracy metrics and methodology are described in [1].
Improved Last-Iterate Convergence of Shuffling Gradient Methods for Nonsmooth Convex Optimization
Authors: Zijian Liu, Zhengyuan Zhou
ICML 2025 | Venue PDF | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Theoretical | We study the convergence of the shuffling gradient method... A recent advance... establishes the first last-iterate convergence results... In this work, we provide the first improved last-iterate analysis... As an important implication, we give the first (nearly) optimal convergence result... We answer the question affirmatively by establishing the first improved last-iterate convergence rates... We prove the following sharper rate... This section describes our key ideas in the analysis, provides the most important Lemma 5.1, and then uses it to sketch the proof. |
| Researcher Affiliation | Collaboration | 1Stern School of Business, New York University 2Arena Technologies. |
| Pseudocode | Yes | Algorithm 1 General Proximal Gradient Method Input: initial point x1 domψ, stepsize ηt > 0, t [T]. for t = 1 to T do Generate an index i(t) [n] xt+1 = argminx Rdψ(x) + fi(t)(xt), x + x xt 2 2ηt Output: x T +1 |
| Open Source Code | No | No explicit statement about providing source code or a link to a repository was found in the paper. |
| Open Datasets | No | The paper is theoretical and focuses on algorithm convergence rates for a general finite-sum function minimization problem. No specific datasets are mentioned or made available for public access. |
| Dataset Splits | No | The paper does not describe experiments using specific datasets, and therefore no dataset split information is provided. |
| Hardware Specification | No | The paper is theoretical and focuses on mathematical analysis of algorithms; it does not describe any computational experiments or specify hardware used. |
| Software Dependencies | No | The paper is theoretical and focuses on mathematical analysis of algorithms; it does not describe any computational experiments or software dependencies with specific version numbers. |
| Experiment Setup | No | The paper focuses on theoretical convergence rates and mathematical properties of algorithms. It does not describe an experimental setup with specific hyperparameters, model initialization, or training configurations. |