Bias and Generalization in Deep Generative Models: An Empirical Study

Authors: Shengjia Zhao, Hongyu Ren, Arianna Yuan, Jiaming Song, Noah Goodman, Stefano Ermon

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

Reproducibility Variable Result LLM Response
Research Type Experimental In this paper we propose a framework to systematically investigate bias and generalization in deep generative models of images. Using this framework, we are able to systematically evaluate generalization patterns of state-of-the-art models such as GAN [5] and VAE [4].
Researcher Affiliation Academia Shengjia Zhao , Hongyu Ren , Arianna Yuan, Jiaming Song, Noah Goodman, Stefano Ermon Stanford University {sjzhao,hyren,xfyuan,tsong,ngoodman,ermon}@stanford.edu
Pseudocode No The paper does not contain any structured pseudocode or algorithm blocks.
Open Source Code Yes Code available at https://github.com/ermongroup/BiasAndGeneralization
Open Datasets Yes We use two different datasets for this experiment: a toy dataset where there are k non-overlapping dots (with random color and location) in the image, as in the numerosity estimation task in cognitive psychology [27, 28], and the CLEVR dataset where there are k objects (with random shape, color, location and size) in the scene [29]. Three MNIST: We use images that contain three MNIST digits.
Dataset Splits No The paper mentions 'training dataset' and 'training set' multiple times and uses large datasets (100k-1M examples), but does not explicitly provide details about training/validation/test splits, percentages, or sample counts.
Hardware Specification No The paper does not provide any specific details about the hardware used to run the experiments, such as GPU or CPU models, or cloud computing specifications.
Software Dependencies No The paper mentions using WGAN-GP [26] and VAE [4] model families, but does not specify any software dependencies with version numbers (e.g., Python, PyTorch, TensorFlow versions, or library versions).
Experiment Setup No The paper mentions using 'different network architectures and hyper-parameter choices', but it does not provide specific details such as learning rates, batch sizes, number of epochs, or optimizer settings in the main text.