Hybrid Heterogeneous Transfer Learning through Deep Learning

Authors: Joey Zhou, Sinno Pan, Ivor Tsang, Yan Yan

AAAI 2014 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Extensive experiments on several multilingual sentiment classification tasks verify the effectiveness of our proposed approach compared with some baseline methods.
Researcher Affiliation Academia Nanyang Technological University, Singapore Institute for Infocomm Research, Singapore University of Technology, Sydney, Australia ]The University of Queensland, Australia
Pseudocode Yes Algorithm 1 Hybrid Heterogeneous Transfer Learning.
Open Source Code No The paper does not provide any concrete access information (e.g., repository link, explicit statement of code release) for the source code of the methodology described.
Open Datasets Yes The cross-language sentiment dataset (Prettenhofer and Stein 2010) comprises of Amazon product reviews of three product categories: books, DVDs and music.
Dataset Splits No The paper specifies a split into 'train file' and 'test file' for the dataset, but does not explicitly mention a 'validation' split or provide details for one.
Hardware Specification No No specific hardware details (e.g., CPU/GPU models, memory, or cloud instance types) used for running experiments were provided in the paper.
Software Dependencies No The paper mentions using 'linear support vector machine (SVM) (Fan et al. 2008)' (referring to LIBLINEAR) and 'm SDA and conduct CCA', but does not provide specific version numbers for any software dependencies.
Experiment Setup Yes Specifically, we choose λ from {0.01, 0.1, 1, 10, 100} for HHTL, choose corruption probability p from {0.5, 0.6, 0.7, 0.8, 0.9} for m SDA from, and fix the number of layers used in m SDA to be 3. We tune the parameter for CL-KCCA (see (5) in (Vinokourov, Shawe Taylor, and Cristianini 2002)), m SDA-CCA, and the parameter β for He Map (See (1) in (Shi et al. 2010)) from {0.01, 0.1, 1, 10, 100}.