Rademacher Complexity Bounds for a Penalized Multi-class Semi-supervised Algorithm
Authors: Yury Maximov, Massih-Reza Amini, Zaid Harchaoui
JAIR 2018 | Venue PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Our theoretical result exhibit convergence rates extending those proposed in the literature for the binary case, and experimental results on different multi-class classification problems show empirical evidence that supports the theory. Section 5. Experimental Results |
| Researcher Affiliation | Academia | Yury Maximov EMAIL Skolkovo Institute of Science and Technology Los Alamos National Laboratory... Massih-Reza Amini EMAIL Universit e Grenoble Alpes CNRS... Zaid Harchaoui EMAIL University of Washington... |
| Pseudocode | Yes | Algorithm 1: Pseudo-code of the PMS2L algorithm |
| Open Source Code | No | The paper does not provide any concrete access to source code for the methodology described. |
| Open Datasets | Yes | We perform experiments on six publicly available datasets. The three first ones are Fungus, Birds and Athletics that consist of three aggregations of lead nodes that go down from parent nodes in the Image Net hierarchy (Russakovsky et al., 2015). The three others collections are respectively the MNIST database of handwritten digits, the pre-processed 20 Newsgroups (20-NG) collection (Chang & Lin, 2011) and the USPS dataset (Hastie, Tibshirani, & Friedman, 2009). |
| Dataset Splits | Yes | The proportions of training and test sets were kept fixed to those given in the released data files. Within the training set (Sℓ Su) we randomly sampled labeled examples Sℓ, with different sizes, and used the remaining as unlabeled data... The parameter C of the SVM classifier is determined by five fold cross-validation... Results are evaluated over the test set using the accuracy, and the reported performance is averaged over 25 random (labeled/unlabeled/test) sets of the initial collections. |
| Hardware Specification | No | The paper does not provide specific hardware details (e.g., CPU/GPU models, memory) used for running its experiments. |
| Software Dependencies | No | The paper mentions using a linear kernel SVM and references LIBSVM (Chang & Lin, 2011), but does not specify the version number of LIBSVM or any other key software components used in their experiments. |
| Experiment Setup | Yes | As the clustering algorithm A, we employed the Nearest Neighbor Clustering technique proposed by Bubeck and Luxburg (2009), and fixed m = 4K, κ = 2 and η = 10 3... The parameter C of the SVM classifier is determined by five fold cross-validation in logarithmic range between 10 4 and 104 over the available labeled training data. |