Density estimation using Real NVP

Authors: Laurent Dinh, Jascha Sohl-Dickstein, Samy Bengio

ICLR 2017 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We demonstrate its ability to model natural images on four datasets through sampling, log-likelihood evaluation, and latent variable manipulations. We train our model on four natural image datasets: CIFAR-10 [36], Imagenet [52], Large-scale Scene Understanding (LSUN) [70], Celeb Faces Attributes (Celeb A) [41].
Researcher Affiliation Collaboration Laurent Dinh Montreal Institute for Learning Algorithms University of Montreal Montreal, QC H3T1J4 Jascha Sohl-Dickstein Google Brain Samy Bengio Google Brain
Pseudocode No The paper includes diagrams illustrating computational flow (e.g., Figure 2) but does not contain any explicitly labeled pseudocode or algorithm blocks.
Open Source Code No The paper does not contain any explicit statements or links indicating that the source code for the methodology described is publicly available. It only mentions 'Tensorflow' in acknowledgments as a tool used.
Open Datasets Yes We train our model on four natural image datasets: CIFAR-10 [36], Imagenet [52], Large-scale Scene Understanding (LSUN) [70], Celeb Faces Attributes (Celeb A) [41].
Dataset Splits Yes We train our model on four natural image datasets: CIFAR-10 [36], Imagenet [52], Large-scale Scene Understanding (LSUN) [70], Celeb Faces Attributes (Celeb A) [41].
Hardware Specification No The paper does not specify any particular hardware components such as GPU or CPU models used for running the experiments.
Software Dependencies No The paper mentions using 'Tensorflow' in the acknowledgments but does not provide a specific version number for it or any other software dependencies.
Experiment Setup Yes For datasets of images of size 32 32, we use 4 residual blocks with 32 hidden feature maps for the first coupling layers with checkerboard masking. Only 2 residual blocks are used for images of size 64 64. We use a batch size of 64. For CIFAR-10, we use 8 residual blocks, 64 feature maps, and downscale only once. We optimize with ADAM [33] with default hyperparameters and use an L2 regularization on the weight scale parameters with coefficient 5 10 5.