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..
Multiwavelet-based Operator Learning for Differential Equations
Authors: Gaurav Gupta, Xiongye Xiao, Paul Bogdan
NeurIPS 2021 | Venue PDF | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We perform experiments on the Korteweg-de Vries (Kd V) equation, Burgers equation, Darcy Flow, and Navier-Stokes equation. |
| Researcher Affiliation | Academia | Gaurav Gupta, Xiongye Xiao, Paul Bogdan Ming Hsieh Department of Electrical and Computer Engineering University of Southern California, Los Angeles, CA 90089 |
| Pseudocode | No | The paper includes a diagram of the model architecture in Figure 2, but it does not provide any structured pseudocode or algorithm blocks. |
| Open Source Code | Yes | The code for reproducing the experiments is available at: https://github.com/gaurav71531/mwt-operator. The code is uploaded with the supplementary materials. |
| Open Datasets | Yes | Unless stated otherwise, the training set is of size 1000 while test is of size 200. A part of the datasets are taken from the FNO work [47], while some are generated using the scripts provided by the same authors. We have properly cited the work in Section 3 Benchmark models. |
| Dataset Splits | No | The paper states the size of the training and test sets ("training set is of size 1000 while test is of size 200") but does not explicitly provide details for a validation set split (e.g., its size or percentage). |
| Hardware Specification | Yes | All of the experiments are performed on a single Nvidia V100 32 GB GPU |
| Software Dependencies | No | The paper mentions the use of 'chebfun package [27]' for numerical solutions but does not provide specific version numbers for this or any other software dependencies crucial for reproducibility. |
| Experiment Setup | Yes | All the models (including ours) are trained for a total of 500 epochs using Adam optimizer with an initial learning rate (LR) of 0.001. The LR decays after every 100 epochs with a factor of γ = 0.5. The loss function is taken as relative L2 error [47]. All of the experiments are performed on a single Nvidia V100 32 GB GPU, and the results are averaged over a total of 3 seeds. |