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..
Generalized Eigenvalue Problems with Generative Priors
Authors: Zhaoqiang Liu, Wen Li, Junren Chen
NeurIPS 2024 | Venue PDF | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Numerical results are provided to demonstrate the effectiveness of the proposed method. |
| Researcher Affiliation | Academia | Zhaoqiang Liu Wen Li University of Electronic Science and Technology of China EMAIL Junren Chen University of Hong Kong EMAIL |
| Pseudocode | Yes | Algorithm 1 Projected Rayleigh Flow Method (PRFM) |
| Open Source Code | No | The code will be open sourced after acceptance. |
| Open Datasets | Yes | In this section, we conduct proof-of-concept numerical experiments on the MNIST dataset [47]... Additional results for MNIST and Celeb A [55] are provided in Appendices D and E. |
| Dataset Splits | No | The paper mentions training on the 'original MNIST training set' and evaluating on a 'testing set', but it does not explicitly specify a validation split percentage or count. |
| Hardware Specification | Yes | All experiments are carried out using Python 3.10.6 and Py Torch 2.0.0, with an NVIDIA RTX 3060 Laptop 6GB GPU. |
| Software Dependencies | Yes | All experiments are carried out using Python 3.10.6 and Py Torch 2.0.0, with an NVIDIA RTX 3060 Laptop 6GB GPU. |
| Experiment Setup | Yes | The VAE is trained using the Adam optimizer with a mini-batch size of 100 and a learning rate of 0.001 on the original MNIST training set. To approximately perform the projection step PG( ), we use a gradient descent method with the Adam optimizer, with a step size of 100 and a learning rate of 0.1. |