AutoReCon: Neural Architecture Search-based Reconstruction for Data-free Compression

Authors: Baozhou Zhu, Peter Hofstee, Johan Peltenburg, Jinho Lee, Zaid Alars

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

Reproducibility Variable Result LLM Response
Research Type Experimental Experimental results show that using generators discovered by the Auto Recon method always improve the performance of data-free compression.
Researcher Affiliation Collaboration 1Delft University of Technology, Delft, The Netherlands 2National University of Defense Technology, Changsha, China 3IBM Austin, Austin, TX, USA 4Yonsei University, Seoul, Korea
Pseudocode Yes Algorithm 1 The Auto Re Conmethod for data-free compression
Open Source Code No The paper does not include an unambiguous statement or a direct link to the source code for the methodology described.
Open Datasets Yes We use the same experimental settings as the GDFQ method... for both CIFAR-100 and Image Net classification
Dataset Splits No The paper mentions 'Lval r and Ltrain r refer to the reconstruction loss function on the reconstructed training dataset and the reconstructed validation dataset, respectively.' but does not provide specific train/validation/test dataset splits (e.g., percentages or sample counts) for reproducibility of the overall experiment.
Hardware Specification No The paper does not explicitly describe the specific hardware used to run its experiments (e.g., specific GPU models, CPU models, or memory details).
Software Dependencies No The paper does not provide specific version numbers for ancillary software components (e.g., Python, PyTorch, TensorFlow versions).
Experiment Setup No The paper states 'We use the same experimental settings as the GDFQ method' but does not explicitly list specific hyperparameter values or training configurations within this paper.