EBEK: Exemplar-Based Kernel Preserving Embedding
Authors: Ahmed Elbagoury, Rania Ibrahim, Mohamed S. Kamel, Fakhri Karray
IJCAI 2016 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | The effectiveness of the proposed approach is evaluated on two tasks, approximate nearest neighbors search and interpretability experiments. Section 4.1 shows the setup and results for the ANNs search task, and section 4.2 shows the interpretability experiments. Figures 2 and 3 show the effect of changing the |T | on the Recall@R and Precision@R for both TDT2 and 20NG datasets. |
| Researcher Affiliation | Academia | University of Waterloo, Waterloo, Ontario, Canada N2L 3G1 {ahmed.elbagoury, rania.ibrahim, mkamel and karray}@uwaterloo.ca |
| Pseudocode | Yes | Algorithm 1 shows the pseudo code of the algorithm. Algorithm 1: Linear Kernel Preserving Embedding. Algorithm 2: get Independentcol: Independent Columns Selection. Algorithm 3: diag Matrix: Diagonalize the Input Matrix. |
| Open Source Code | No | The paper does not provide any explicit statement about making the source code available or a link to a code repository for the methodology described. |
| Open Datasets | Yes | We have used four datasets, COIL20 which contains 1440 samples in 1024 dimensional space, ISOLET which contains 1560 samples in 617 dimensions, TDT2 which contains 9394 sample in 19677 dimensional space and a subset of 20 Newsgroups (20NG in short) containing 9990 samples in 29360 dimensional space [Cai et al., 2009]. |
| Dataset Splits | No | The paper describes the datasets used and how search quality is measured using Recall@R and Precision@R, but it does not specify training, validation, or test dataset splits in percentages, absolute counts, or reference predefined splits for reproducing the data partitioning for model training. |
| Hardware Specification | No | The paper does not provide any specific hardware details such as GPU or CPU models, memory, or cloud computing instances used for running the experiments. |
| Software Dependencies | No | The paper does not provide specific software dependencies or library versions used for the experiments (e.g., Python, PyTorch, TensorFlow, scikit-learn versions). |
| Experiment Setup | Yes | Additionally, the number of the basis in the lower dimensional space m is set to 10 in all the techniques. Note that the results in this subsection are not affected by the value of as discussed in section 3.2. ... The value of was chosen empirically to yield the best visualization results and was set to 0.65 and 0.94 in COIL20 and ORL datasets respectively. |