Scalable Gaussian Process Regression Using Deep Neural Networks

Authors: Wenbing Huang, Deli Zhao, Fuchun Sun, Huaping Liu, Edward Chang

IJCAI 2015 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental For the experiments on three representative large datasets, our proposed models significantly outperform the state-of-the-art algorithms of Gaussian process regression. 5 Experiments We compare the performance of proposed models with full GPs, FITCs [Snelson and Ghahramani, 2005] and mix MNNs [L azaro-Gredilla and Figueiras-Vidal, 2010] on one synthetic image dataset Rectangles [Larochelle et al., 2007] and two real face datasets Olivetti Faces [Salakhutdinov and Hinton, 2007] and FGnet [Zhou et al., 2005].
Researcher Affiliation Collaboration Wenbing Huang1, Deli Zhao2, Fuchun Sun1, Huaping Liu1, Edward Chang2 1State Key Laboratory of Intelligent Technology and System, Tsinghua University, Beijing, China 2HTC Research, Beijing, China
Pseudocode No The paper describes the methodology using text and mathematical equations, but does not include any structured pseudocode or algorithm blocks.
Open Source Code No The paper does not provide a statement about releasing open-source code or a direct link to a code repository for the methodology described.
Open Datasets Yes on one synthetic image dataset Rectangles [Larochelle et al., 2007] and two real face datasets Olivetti Faces [Salakhutdinov and Hinton, 2007] and FGnet [Zhou et al., 2005].
Dataset Splits No Table 1 lists '#Train' and '#Test' samples for each dataset, but the paper does not explicitly provide details about a validation dataset split or how it was used to reproduce experiments.
Hardware Specification Yes All experiments are carried out with Matlab 8.1.0.604 (R2013a) on Intel Core i7, 2.90-GHz CPU with 8-GB RAM.
Software Dependencies Yes All experiments are carried out with Matlab 8.1.0.604 (R2013a) on Intel Core i7, 2.90-GHz CPU with 8-GB RAM.
Experiment Setup Yes The setting of training SDAEs in DNN-GPs is fixed as follows: The corrupted proportion v in each DAE is set to be 0.5. Equation (7) is optimized with the gradient descent method whose learning rate is 0.1. The dataset is divided into mini-batches of size 100, and the parameters are updated after each minibatch iteration. Each DAE is trained for 30 passes through the entire training set. If the objective function dose not decrease within 50 evaluations in each line search, or the epoch of the line search exceeds 100, the training process will halts.