The Elastic Lottery Ticket Hypothesis

Authors: Xiaohan Chen, Yu Cheng, Shuohang Wang, Zhe Gan, Jingjing Liu, Zhangyang Wang

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

Reproducibility Variable Result LLM Response
Research Type Experimental We conduct extensive experiments on CIFAR-10 and Image Net, and propose a variety of strategies to tweak the winning tickets found from different networks of the same model family (e.g., Res Nets).
Researcher Affiliation Collaboration Xiaohan Chen1 Yu Cheng2 Shuohang Wang2 Zhe Gan2 Jingjing Liu3 Zhangyang Wang1 1University of Texas at Austin 2Microsoft Corporation 3Tsinghua University
Pseudocode No No structured pseudocode or algorithm blocks found. The IMP algorithm steps are described in paragraph form.
Open Source Code Yes Code is available at https://github.com/VITA-Group/Elastic LTH.
Open Datasets Yes We conduct extensive experiments on CIFAR-10 [26] and then Image Net [7]
Dataset Splits Yes We conduct extensive experiments on CIFAR-10 [26] and then Image Net [7], transferring the winning tickets across multiple models from Res Net family and VGG family.
Hardware Specification No No specific hardware details (e.g., exact GPU/CPU models, memory amounts) are mentioned for running experiments.
Software Dependencies No No specific ancillary software details (e.g., library or solver names with version numbers) are mentioned in the provided text.
Experiment Setup No No specific experimental setup details (e.g., concrete hyperparameter values, training configurations, or system-level settings) are explicitly provided in the main text of the paper.