Variational Diffusion Models

Authors: Diederik Kingma, Tim Salimans, Ben Poole, Jonathan Ho

NeurIPS 2021 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We introduce a flexible family of diffusion-based generative models that achieve new state-of-the-art log-likelihoods on standard image density estimation benchmarks (CIFAR-10 and Image Net)... We empirically compare to these works, as well as others, in Table 1.
Researcher Affiliation Industry Google Research
Pseudocode No The paper does not contain any clearly labeled pseudocode or algorithm blocks.
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 demonstrate our proposed class of diffusion models, which we call Variational Diffusion Models (VDMs), on the CIFAR-10 [Krizhevsky et al., 2009] dataset, and the downsampled Image Net [Van Oord et al., 2016, Deng et al., 2009] dataset, where we focus on maximizing likelihood.
Dataset Splits No The paper mentions training and testing on datasets like CIFAR-10 and ImageNet, but it does not explicitly provide details about specific train/validation/test splits (e.g., percentages or sample counts) needed for reproduction beyond mentioning the 'test set'.
Hardware Specification Yes Our models are trained on NVIDIA A100 GPUs...
Software Dependencies No The paper states, 'We implemented our models in JAX [Bradbury et al., 2018] and Haiku [Anonymous, 2021],' but it does not provide specific version numbers for these software dependencies.
Experiment Setup Yes We use an Adam optimizer [Kingma and Ba, 2014] with 10−4 learning rate. We train for 2000K steps with a batch size of 2048. More details of our model specifications are in Appendix B.