Closing the Dequantization Gap: PixelCNN as a Single-Layer Flow

Authors: Didrik Nielsen, Ole Winther

NeurIPS 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Finally, we study multilayer flows composed of Pixel CNNs and non-autoregressive coupling layers and demonstrate state-of-the-art results on CIFAR-10 for flow models trained with dequantization. 6 Experiments
Researcher Affiliation Academia Didrik Nielsen Technical University of Denmark didni@dtu.dk Ole Winther Technical University of Denmark olwi@dtu.dk
Pseudocode No No structured pseudocode or algorithm blocks were found in the paper.
Open Source Code Yes The code used for experiments is publicly available at https://github.com/ didriknielsen/pixelcnn_flow.
Open Datasets Yes We trained Pixel CNN (van den Oord et al., 2016c) and Pixel CNN++ (Salimans et al., 2017) as flow models on CIFAR-10
Dataset Splits No The paper mentions using CIFAR-10, which has standard splits, but does not explicitly provide the train/validation/test split percentages or sample counts within the text.
Hardware Specification No The paper does not provide specific hardware details (e.g., exact GPU/CPU models, processor types, or memory amounts) used for running its experiments.
Software Dependencies No The paper does not provide specific ancillary software details (e.g., library or solver names with version numbers) needed to replicate the experiment.
Experiment Setup No The paper describes the architectural composition of models and the dequantization method used, but does not provide specific hyperparameter values (e.g., learning rate, batch size, number of epochs, optimizer settings) or a detailed table/paragraph of training configurations.