Self-Paced Two-dimensional PCA

Authors: Jiangxin Li, Zhao Kang, Chong Peng, Wenyu Chen8392-8400

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

Reproducibility Variable Result LLM Response
Research Type Experimental Extensive experiments on image reconstruction and clustering verify the superiority of our approach. Four benchmark image databases, including ORL, MNIST, AR and EYale B, are utilized in the experiments.
Researcher Affiliation Academia Jiangxin Li,1 Zhao Kang, 1 Chong Peng, 2 Wenyu Chen1 1 School of Computer Science and Engineering, University of Electronic Science and Technology of China 2 College of Computer Science and Technology, Qingdao University
Pseudocode Yes Algorithm 1 SP2DPCA
Open Source Code Yes The code of our implementation is published 1. 1https://github.com/sckangz/SP2DPCA.
Open Datasets Yes Four benchmark image databases, including ORL, MNIST, AR and EYale B, are utilized in the experiments.
Dataset Splits No Following the comparison methods, half of the images are used for training and the rest is left for testing. The paper mentions train and test splits but does not specify a separate validation split.
Hardware Specification No The paper mentions running times but does not provide specific hardware details (e.g., CPU/GPU models, memory) used for the experiments.
Software Dependencies No The paper mentions using 'k-means clustering algorithm' but does not provide version numbers for any software dependencies.
Experiment Setup Yes For SP2DPCA, the optimal parameters are searched in the range ΞΆ = {50, 100, 200, 500, 1000} and c = {300, 500, 1000, 3000, 5000} and the best results are recorded correspondingly. We stop the algorithm when the loss value does not change much.