Semi-supervised Orthogonal Graph Embedding with Recursive Projections

Authors: Hanyang Liu, Junwei Han, Feiping Nie

IJCAI 2017 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Comprehensive experiment on several benchmarks demonstrates the significant improvement over the existing methods.
Researcher Affiliation Academia Hanyang Liu1, Junwei Han1 , Feiping Nie1,2 1 Northwestern Polytechnical University, Xi an 710072, P. R. China 2 University of Texas at Arlington, USA
Pseudocode Yes Algorithm 1 Algorithm to solve problem in Eq.(16)
Open Source Code No The paper does not provide concrete access to source code for the described methodology.
Open Datasets Yes In our experiments, we use six real world benchmarks including three face benchmarks (JAFFE1, AT&T2, and CMU-PIE), a handwritten digits dataset MNIST, and two object benchmarks (COIL-20 and MPEG73). 1http://www.kasrl.org/jaffe.html 2http://www.cl.cam.ac.uk/research/dtg/attarchive.html 3http://www.dabi.temple.edu/ shape/MPEG7/dataset.html
Dataset Splits No The paper describes training and testing splits and labeled/unlabeled data, but does not explicitly mention a separate validation dataset split.
Hardware Specification No The paper does not provide specific hardware details (e.g., CPU, GPU models, memory) used for running the experiments.
Software Dependencies No The paper does not provide specific ancillary software details with version numbers.
Experiment Setup Yes In SOGE, we set the weight µ in the diagonal matrix U as 100 for all datasets. In order to fairly compare SOGE with other algorithms, we tuned all the regularization parameters involved in each algorithms with grid search within {10 9, 10 6, 10 3, 100, 103, 106, 109}. For all the algorithms, we employ the k-nearest neighbor (k NN) classifier to evaluate the performance of dimensionality reduction, and set k = 1 in k NN for all the algorithms. For all the datasets, we use PCA as a preprocessing procedure to denoise all the data with 95% of the information preserved, similarly as in [Yan et al., 2007].