Bidirectional Convolutional Poisson Gamma Dynamical Systems

Authors: wenchao chen, Chaojie Wang, Bo Chen, Yicheng Liu, Hao Zhang, Mingyuan Zhou

NeurIPS 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experimental results on a variety of document corpora demonstrate that the proposed models can extract expressive multi-level latent representations, including interpretable phrase-level topics and sentence-level temporal transitions as well as discriminative document-level features, achieving state-of-the-art document categorization performance while being memory and computation efficient.
Researcher Affiliation Academia Wenchao Chen , Chaojie Wang , Bo Chen , Yicheng Liu, Hao Zhang National Laboratory of Radar Signal Processing Xidian University, Xi an, Shaanxi 710071, China wcchen_xidian@163.com, xd_silly@163.com, bchen@mail.xidian.edu.cn, moooooore66@gmail.com, zhanghao_xidian@163.com Mingyuan Zhou Mc Combs School of Business The University of Texas at Austin Austin, TX 78712, USA mingyuan.Zhou@mccombs.utexas.edu
Pseudocode No The paper describes mathematical formulations and generative processes but does not include explicit pseudocode blocks or algorithms labeled as such.
Open Source Code Yes Python (Py Torch) code is provided at https://github.com/Bo Chen Group/BCPGDS.
Open Datasets Yes We evaluate the effectiveness of our model on five large corpora, including ELEC, IMDB-2, Reuters, Yelp 2014, and IMDB-10, which are described in detail in the Appendix.
Dataset Splits Yes We evaluate our model with different amount of labeled data (5%, 10%, 25%, or 50%), besides the whole training set being unlabeled data.
Hardware Specification Yes Table 3: Comparison of testing times (seconds) with batch-size 128 on two RTX 2080 Ti GPUs.
Software Dependencies No The paper mentions "Python (Py Torch) code is provided" but does not specify version numbers for Python, PyTorch, or other relevant libraries.
Experiment Setup Yes We fix the hyperparameters of our models as τ0 = 1, ϵ0 = 0.1, γ0 = 0.1, η = 0.05 for all experiments.