Quantitative Performance Assessment of CNN Units via Topological Entropy Calculation

Authors: Yang Zhao, Hao Zhang

ICLR 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental 4 EXPERIMENTS
Researcher Affiliation Academia Yang Zhao & Hao Zhang Department of Electronic Engineering Tsinghua University zhao-yan18@mails.tsinghua.edu.cn, haozhang@tsinghua.edu.cn
Pseudocode No The paper describes computational steps and flow, but does not include any explicitly labeled pseudocode or algorithm blocks.
Open Source Code No The paper does not provide any explicit statement or link regarding the availability of open-source code for the described methodology.
Open Datasets Yes For experiments, we use the VGG16 network architecture to perform the image classification task on the Image Net dataset.
Dataset Splits No The paper refers to 'training set' and 'test set' but does not explicitly provide details about a validation set or specific split percentages for reproduction.
Hardware Specification Yes All implementations are deployed on the Nvidia-A100 station with a batch size of 512.
Software Dependencies No The paper mentions using an 'SGD optimizer' but does not provide specific software names with version numbers for reproducibility (e.g., Python, PyTorch/TensorFlow versions, or other libraries).
Experiment Setup Yes The hyper-parameters are the same with them in the paper (Simonyan & Zisserman, 2014). Also, similar to the setting in the paper, all the images are simply resized to 224 224.