Variational Autoencoders with Jointly Optimized Latent Dependency Structure

Authors: Jiawei He, Yu Gong, Joseph Marino, Greg Mori, Andreas Lehrmann

ICLR 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We validate our framework in extensive experiments on MNIST, Omniglot, and CIFAR-10. Comparisons to state-of-the-art structured variational autoencoder baselines show improvements in terms of the expressiveness of the learned model.
Researcher Affiliation Academia Jiawei He1 & Yu Gong1 {jha203, gongyug}@sfu.ca Joseph Marino2 jmarino@caltech.edu Greg Mori1 mori@cs.sfu.ca Andreas M. Lehrmann andreas.lehrmann@gmail.com 1School of Computing Science, Simon Fraser University Burnaby, BC, V5B1Z1, Canada 2California Institute of Technology Pasadena, CA, 91125, USA
Pseudocode Yes Algorithm 1 Optimizing VAEs with Latent Dependency Structure
Open Source Code No The paper does not provide an explicit statement or link indicating that the source code for the described methodology is publicly available.
Open Datasets Yes We evaluate the proposed latent dependency learning approach on three benchmark datasets: MNIST (Lecun et al., 1998; Larochelle & Murray, 2011), Omniglot (Lake et al., 2013), and CIFAR10 (Krizhevsky, 2009).
Dataset Splits No The paper mentions using benchmark datasets and evaluating on a 'test set' but does not explicitly provide details about training, validation, and test dataset splits (e.g., percentages, sample counts, or predefined split citations for all three).
Hardware Specification No The paper does not provide specific details about the hardware (e.g., GPU/CPU models, memory) used to run the experiments.
Software Dependencies No The paper mentions 'Py Torch (Paszke et al., 2017)' and 'Adam (Kingma & Ba, 2015)', but does not provide specific version numbers for these software components to ensure reproducibility.
Experiment Setup Yes All models were implemented with Py Torch (Paszke et al., 2017) and trained using the Adam (Kingma & Ba, 2015) optimizer with a mini-batch size of 64 and learning rate of 1e 3. Learning rate is decresed by 0.25 every 200 epochs. The Gumbel-softmax temperature was initialized at 1 and decreased to 0.99epoch at each epoch. MNIST and Omniglot took 2000 epochs to converge, and CIFAR took 3000 epochs to converge.