Online Heterogeneous Transfer Metric Learning

Authors: Yong Luo, Tongliang Liu, Yonggang Wen, Dacheng Tao

IJCAI 2018 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experiments in diverse applications demonstrate both effectiveness and efficiency of the proposed method.
Researcher Affiliation Academia 1 School of Computer Science and Engineering, Nanyang Technological University, Singapore 2 UBTECH Sydney AI Centre, SIT, FEIT, University of Sydney, Australia
Pseudocode Yes Algorithm 1 The proposed online heterogeneous transfer metric learning (OHTML) algorithm.
Open Source Code No The paper does not provide any explicit statements about the release of its source code or links to a code repository.
Open Datasets Yes This set of experiments is conducted on the popular Caltech101 [Fei-Fei et al., 2004] dataset... We conduct scene clustering on the Scene-15 [Lazebnik et al., 2006] dataset... we employ the well-known labeled face in the wild (LFW) [Huang et al., 2007] dataset... and the Corel5K [Duygulu et al., 2002] dataset is used for evaluation.
Dataset Splits No For the Caltech-101 dataset: "Half of the dataset is used for training and the remaining is for test." For Scene-15: "We randomly split the dataset into equal size for training and test." For LFW: "We adopt the standard 10-folds split of the dataset [Huang et al., 2007], and each fold is used for test in turn." For Corel5K: "half of the data are used for training." While hyperparameters are tuned, the paper does not specify a distinct validation set split for this tuning for all datasets or in general.
Hardware Specification Yes The algorithms are implemented using Matlab and the experiments are conducted on a 3.4 GHz Intel Xeon E5-2687W (8 cores) computer.
Software Dependencies No The algorithms are implemented using Matlab. The paper mentions Matlab but does not provide a specific version number, nor does it list any other software dependencies with version numbers.
Experiment Setup Yes The candidate set for choosing the trade-off hyper-parameters is {10i|i = 5, , 4} if unspecified in the original papers. The hyper-parameters γA and γI are tuned on the set {10i|i = 5, , 4} and {10i|i = 2, , 7} respectively. ... We run the experiments ten times by randomly choosing the labeled set.