Mixed batches and symmetric discriminators for GAN training

Authors: Thomas LUCAS, Corentin Tallec, Yann Ollivier, Jakob Verbeek

ICML 2018 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experimentally, our approach reduces mode collapse in GANs on two synthetic datasets, and obtains good results on the CIFAR10 and Celeb A datasets, both qualitatively and quantitatively.
Researcher Affiliation Collaboration 1Universit e Grenoble Alpes, Inria, CNRS, Grenoble INP, LJK, 38000 Grenoble, France. 2Universit e Paris Sud, INRIA, equipe TAU, Gif-sur-Yvette, 91190, France. 3Facebook Artificial Intelligence Research Paris, France.
Pseudocode No The paper describes the architecture and methods in text and figures but does not contain structured pseudocode or algorithm blocks.
Open Source Code No The paper does not provide concrete access to source code for the methodology described, nor does it include a specific repository link or an explicit code release statement.
Open Datasets Yes Experimentally, our approach reduces mode collapse in GANs on two synthetic datasets, and obtains good results on the CIFAR10 and Celeb A datasets, both qualitatively and quantitatively. and The synthetic dataset from Zhang et al. (2017) is explicitly designed to test mode dropping.
Dataset Splits No The paper does not provide specific dataset split information (exact percentages, sample counts, citations to predefined splits, or detailed splitting methodology) needed to reproduce the data partitioning.
Hardware Specification No The paper does not provide specific hardware details (exact GPU/CPU models, processor types with speeds, memory amounts, or detailed computer specifications) used for running its experiments.
Software Dependencies No The paper mentions optimizers and network types but does not provide specific ancillary software details, such as library or solver names with version numbers, needed to replicate the experiment.
Experiment Setup Yes The models are trained on their respective losses using the Adam (Kingma & Ba, 2015) optimizer, with default parameters. The discriminator is trained for five steps for each generator step. and The same Adam hyperparameters from (Miyato et al., 2018) are used for all models: α = 2e 4, β1 = 0.5, β2 = 0.999, and no learning rate decay. We performed hyperparameter search for the number of discrimination steps between each generation step, ndisc, over the range {1, . . . , 5}, and for the batch smoothing parameter γ over [0.2, 0.5]. All models are trained for 400,000 iterations, counting both generation and discrimination steps.