Provably Faster Algorithms for Bilevel Optimization

Authors: Junjie Yang, Kaiyi Ji, Yingbin Liang

NeurIPS 2021 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Our experiments validate our theoretical results and demonstrate the superior empirical performance of our algorithms in hyperparameter applications.
Researcher Affiliation Academia Junjie Yang Department of ECE The Ohio State University yang.4972@osu.edu Kaiyi Ji Department of EECS University of Michigan kaiyiji@umich.edu Yingbin Liang Department of ECE The Ohio State University liang.889@osu.edu
Pseudocode Yes Algorithm 1 Momentum-based Recursive Bilevel Optimizer (MRBO) Algorithm 2 Variance Reduction Bilevel Optimizer (VRBO)
Open Source Code Yes Our codes are available online at https://github.com/JunjieYang97/MRVRBO. Our code is public on Git Hub.
Open Datasets Yes Our experiments are run over a hyper-cleaning application on MNIST. We specify that the dataset we use are public in Appendix B.
Dataset Splits Yes Did you specify all the training details (e.g., data splits, hyperparameters, how they were chosen)? [Yes] The experimental details are specified in Appendix B.
Hardware Specification Yes Did you include the total amount of compute and the type of resources used (e.g., type of GPUs, internal cluster, or cloud provider)? [Yes] The details are in included in Appendix B.
Software Dependencies No The paper does not explicitly list software dependencies with specific version numbers in the main text or refer to such details being in the appendix.
Experiment Setup Yes Did you specify all the training details (e.g., data splits, hyperparameters, how they were chosen)? [Yes] The experimental details are specified in Appendix B.