Hypernetwork-based Meta-Learning for Low-Rank Physics-Informed Neural Networks

Authors: Woojin Cho, Kookjin Lee, Donsub Rim, Noseong Park

NeurIPS 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental 4 Experiments We demonstrate that our proposed method significantly outperforms baselines on the 1-dimensional/2dimensional PDE benchmarks that are known to be very challenging for PINNs to learn [17, 18].
Researcher Affiliation Academia Woojin Cho Kookjin Lee Donsub Rim Noseong Park Yonsei University Arizona State University Washington University in St. Louis snowmoon@yonsei.ac.kr, kookjin.lee@asu.edu, rim@wustl.edu, noseong@yonsei.ac.kr
Pseudocode No The paper mentions 'See Appendix E for the formal algorithm' but Appendix E is not provided in the current document. No pseudocode or algorithm blocks are present in the main body.
Open Source Code No The paper refers to 'Appendix F, including hyperparameter configuration and software/hardware environments' for reproducibility, but does not provide an explicit statement about open-source code release or a link to a repository for the described methodology.
Open Datasets No The paper describes solving partial differential equations (PDEs) by training on generated collocation points, rather than using a publicly available dataset with specific access information.
Dataset Splits No The paper describes a two-phase training process (Phase 1 for meta-learning, Phase 2 for fine-tuning on test PDE parameters) and uses 'test collocation points' for evaluation, but does not specify distinct training, validation, and test dataset splits with percentages or counts.
Hardware Specification No The paper mentions 'Appendix F, including hyperparameter configuration and software/hardware environments' for reproducibility, but Appendix F is not provided in the current document, and no specific hardware details are mentioned elsewhere.
Software Dependencies No The paper mentions 'Appendix F, including hyperparameter configuration and software/hardware environments' for reproducibility, but Appendix F is not provided in the current document, and no specific software dependencies with version numbers are mentioned elsewhere.
Experiment Setup No The paper refers to 'Appendix F, including hyperparameter configuration and software/hardware environments' for reproducibility, but Appendix F is not provided in the current document, and no specific experimental setup details or hyperparameter values are mentioned elsewhere.