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 Coakley et alK. L. Coakley, T. Snelleman, H. Hoos, and O. E. Gundersen, "The embrace of open science: An analysis of a decade of AI research and 56 800 conference papers," Under Review, 2026..
Non-convex online learning via algorithmic equivalence
Authors: Udaya Ghai, Zhou Lu, Elad Hazan
NeurIPS 2022 | Venue PDF | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We empirically verify that reparameterized GD iterates and EG iterates stay close on a toy problem in Figure 1. |
| Researcher Affiliation | Collaboration | Google AI Princeton Princeton University |
| Pseudocode | Yes | Algorithm 1 Online Mirror Descent; Algorithm 2 Online Gradient Descent |
| Open Source Code | No | The paper does not provide any explicit statements about releasing source code or links to a code repository for the methodology described. |
| Open Datasets | No | The paper discusses theoretical examples like 'Exponentiated gradient using quadratic reparameterization' and a 'toy problem' for empirical verification, but it does not specify or provide access information for any publicly available or open datasets used in a typical experimental setup. |
| Dataset Splits | No | The paper is primarily theoretical and does not mention specific training, validation, or test dataset splits. |
| Hardware Specification | No | The paper does not provide any specific details about the hardware used to run the experiments. |
| Software Dependencies | No | The paper does not provide specific software dependencies with version numbers needed to replicate the experiment. |
| Experiment Setup | No | The paper is theoretical and does not provide specific experimental setup details such as hyperparameter values, model initialization, or training schedules. While it defines a parameter η for the regret bound, this is part of the theoretical analysis, not an experimental setup for a training process. |