Co-Regularized PLSA for Multi-Modal Learning

Authors: Xin Wang, MingChing Chang, Yiming Ying, Siwei Lyu

AAAI 2016 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We evaluate the performance of the co PLSA algorithms on text/image cross-modal retrieval tasks, on which they show competitive performance with state-of-the-art methods.
Researcher Affiliation Academia 1Department of Computer Science, 2Department of Mathematics and Statistics University at Albany, State University of New York Albany, NY 12222
Pseudocode Yes The right panel of Fig.1 provides the pseudo-code of the overall algorithm.
Open Source Code No The paper does not provide any statement or link indicating that the source code for the described methodology is publicly available.
Open Datasets Yes We use two benchmark text/image datasets in our experiments: TVGraz (Khan, Saffari, and Bischof 2009) and Wikipedia (Rasiwasia et al. 2010)
Dataset Splits No The paper mentions 'choose the balance parameter λ in co PLSA algorithms with cross-validation on a subset of the training data,' but it does not specify a distinct validation split (e.g., 80/10/10) for evaluating the model during training.
Hardware Specification No The paper does not provide any specific details about the hardware (e.g., CPU, GPU models, memory) used to run the experiments.
Software Dependencies No The paper mentions 'Lambert W function... is provided in popular numerical tools such as MATLAB (function lambertw) or Sci Py (function scipy.special.lambertw),' but it does not provide specific version numbers for MATLAB or SciPy.
Experiment Setup Yes We extract 50 topics from the TVGraz dataset and 100 topics from the Wikipedia dataset, and choose the balance parameter λ in co PLSA algorithms with cross-validation on a subset of the training data.