Aggregating Crowd Wisdoms with Label-aware Autoencoders

Authors: Li'ang Yin, Jianhua Han, Weinan Zhang, Yong Yu

IJCAI 2017 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experiments on three real-world datasets show that proposed models achieve impressive inference accuracy improvement over state-of-the-art models.
Researcher Affiliation Academia Li ang Yin, Jianhua Han, Weinan Zhang, Yong Yu Shanghai Jiao Tong University No.800 Dongchuan Road Shanghai, 200240, China {yinla,hanjianhua44,wnzhang,yyu}@apex.sjtu.edu.cn
Pseudocode No The paper describes the model with mathematical equations and architectural diagrams but does not include structured pseudocode or algorithm blocks.
Open Source Code No The paper mentions implementing models with TensorFlow but does not explicitly state that their own source code for the methodology is available or provide a link to it.
Open Datasets Yes Bluebirds [Welinder et al., 2010]... Flowers [Tian and Zhu, 2015b]... Web Search [Zhou et al., 2012]
Dataset Splits No A dataset is split into training set and validation set. Training process stops when the loss on the validation set begins to increase. The paper mentions the use of training and validation sets, but does not provide specific details on the split percentages, sample counts, or the methodology used for splitting (e.g., random seed, stratified splitting, or k-fold cross-validation).
Hardware Specification No The paper only mentions 'GPU acceleration' for TensorFlow implementation but provides no specific hardware details such as GPU/CPU models, memory, or specific computing environments.
Software Dependencies No The paper mentions using 'TensorFlow' for implementation but does not specify a version number or other software dependencies with their versions.
Experiment Setup Yes Hyperparameters include learning rate η [0.001, 0.1], constraint strength λkl [0.0001, 0.1], and λs [0.0001, 0.1]. We grid-search proper hyperparameters... Gradient descent is exploited to minimize the loss. For LAA-L, we set the number of latent aspects as 2. In this paper, we implement networks with one layer for the classifier and the reconstructor respectively.