Regularizing Variational Autoencoder with Diversity and Uncertainty Awareness

Authors: Dazhong Shen, Chuan Qin, Chao Wang, Hengshu Zhu, Enhong Chen, Hui Xiong

IJCAI 2021 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental extensive experiments on three benchmark datasets clearly show that our approach can outperform state-of-the-art baselines on both likelihood estimation and underlying classification tasks.
Researcher Affiliation Collaboration 1 School of Computer Science and Technology, University of Science and Technology of China 2Baidu Talent Intelligence Center 3Rutgers, The State University of New Jersey
Pseudocode Yes Algorithm 1 Training Procedure of DU-VAE
Open Source Code No The paper does not provide an explicit statement or link for open-source code.
Open Datasets Yes we evaluated our method on two text benchmark datasets, i.e., Yahoo and Yelp corpora [Yang et al., 2017] and one image benchmark dataset, i.e., OMNIGLOT [Lake et al., 2015].
Dataset Splits No The paper mentions using 'dynamically binarized images for training and the fixed binarization as test data' for OMNIGLOT, but does not specify a validation split or provide specific numeric splits for all datasets.
Hardware Specification No The paper does not provide specific hardware details (e.g., GPU/CPU models, memory) used for running experiments.
Software Dependencies No The paper mentions optimizers (SGD, Adam) and model architectures (LSTM, ResNet, Pixel CNN) but does not provide specific software names with version numbers for reproducibility (e.g., PyTorch version, Python version).
Experiment Setup Yes For text datasets, we utilized a single layer LSTM as both encoder and decoder networks... For images, a 3-layer Res Net [He et al., 2016] encoder and a 13-layer Gated Pixel CNN [Van den Oord et al., 2016] decoder are applied. We set the dimension of z as 32. and utilized SGD to optimize the ELBO objective for text and Adam [Kingma and Ba, 2015] for images. Following [Bowman et al., 2015a], we applied a linear annealing strategy to increasing the KL weight from 0 to 1 in the first 10 epochs if possible.