Convolutional Poisson Gamma Belief Network
Authors: Chaojie Wang, Bo Chen, Sucheng Xiao, Mingyuan Zhou
ICML 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Experimental results demonstrate that CPGBN can extract high-quality text latent representations that capture the word order information, and hence can be leveraged as a building block to enrich a wide variety of existing latent variable models that ignore word order. We test the proposed CPGBN and its supervised extension (s CPGBN) on various benchmarks, including: MR: Movie reviews with one sentence per review, where the task is to classify a review as being positive or negative (Pang & Lee, 2005). |
| Researcher Affiliation | Academia | 1National Laboratory of Radar Signal Processing, Collaborative Innovation Center of Information Sensing and Understanding, Xidian University, Xi an, Shaanxi, China. 2Mc Combs School of Business, The University of Texas at Austin, Austin, Texas 78712, USA. Correspondence to: Bo Chen <bchen@mail.xidian.edu.cn>. |
| Pseudocode | Yes | Algorithm 1 Hybrid stochastic-gradient MCMC and autoencoding variational inference for CPGBN |
| Open Source Code | No | The paper states that the algorithm is implemented in Tensor Flow and py CUDA but does not provide any explicit statement about releasing the source code for their method or a link to a repository. |
| Open Datasets | Yes | MR: Movie reviews with one sentence per review, where the task is to classify a review as being positive or negative (Pang & Lee, 2005). TREC: TREC question dataset, where the task is to classify a question into one of six question types (whether the question is about abbreviation, entity, description, human, location, or numeric) (Li & Roth, 2002). SUBJ: Subjectivity dataset... ELEC: ELEC dataset (Mcauley & Leskovec, 2013) consists of electronic product reviews... IMDB: IMDB dataset (Maas et al., 2011) is a benchmark dataset for sentiment analysis... |
| Dataset Splits | Yes | Test: Test set size, where CV means 10-fold cross validation. On each network, we apply the Gibbs sampler to collect 200 MCMC samples after 500 burn-ins to estimate the posterior mean of the feature usage weight vector wj, for every document in both the training and testing sets. |
| Hardware Specification | Yes | Note all running times are reported based on a Nvidia GTX 1080Ti GPU. |
| Software Dependencies | No | The paper mentions that the algorithm is implemented in 'Tensor Flow (Abadi et al., 2016)' and 'py CUDA (Klockner et al., 2012)' but does not specify exact version numbers for these software dependencies. |
| Experiment Setup | Yes | Under the same configuration of filter width F = 3 and the same hyperparameter setting, where e0 = f0 = 0.1 and η(t) = 0.05, the networks are trained with the proposed Gibbs sampler. To make a fair comparison, we let CPGBNs to have the same hidden layer widths as the other methods, and set the filter width as 3 for the convolutional layer. |