Deep Unsupervised Learning using Nonequilibrium Thermodynamics

Authors: Jascha Sohl-Dickstein, Eric Weiss, Niru Maheswaranathan, Surya Ganguli

ICML 2015 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We demonstrate the utility of these diffusion probabilistic models by training high log likelihood models for a two-dimensional swiss roll, binary sequence, handwritten digit (MNIST), and several natural image (CIFAR-10, bark, and dead leaves) datasets. ... 3. Experiments We train diffusion probabilistic models on a variety of continuous datasets, and a binary dataset. We then demonstrate sampling from the trained model and inpainting of missing data, and compare model performance against other techniques.
Researcher Affiliation Academia Jascha Sohl-Dickstein JASCHA@STANFORD.EDU Stanford University Eric A. Weiss EWEISS@BERKELEY.EDU University of California, Berkeley Niru Maheswaranathan NIRUM@STANFORD.EDU Stanford University Surya Ganguli SGANGULI@STANFORD.EDU Stanford University
Pseudocode No No structured pseudocode or algorithm blocks are provided.
Open Source Code Yes We additionally release an open source reference implementation of the algorithm. ... A reference implementation of the algorithm utilizing Blocks (van Merri enboer et al., 2015) is available at https://github.com/Sohl-Dickstein/Diffusion-Probabilistic-Models.
Open Datasets Yes We demonstrate the utility of these diffusion probabilistic models by training high log likelihood models for a two-dimensional swiss roll, binary sequence, handwritten digit (MNIST), and several natural image (CIFAR-10, bark, and dead leaves) datasets. ... CIFAR-10 (Krizhevsky & Hinton, 2009) dataset. ... MNIST digits (Le Cun & Cortes, 1998). ... Dead leaf images (Jeulin, 1997; Lee et al., 2001). ... bark texture images (T01-T04) from (Lazebnik et al., 2005).
Dataset Splits No The lower bound K on the log likelihood, computed on a holdout set, for each of the trained models.
Hardware Specification No No specific hardware details (e.g., GPU/CPU models, memory) are provided for the experimental setup.
Software Dependencies No In all cases the objective function and gradient were computed using Theano (Bergstra & Breuleux, 2010), and model training was with SFO (Sohl-Dickstein et al., 2014). A reference implementation of the algorithm utilizing Blocks (van Merri enboer et al., 2015) is available at...
Experiment Setup Yes For all results in this paper, multi-layer perceptrons are used to define these functions. ... The multi-scale convolutional architecture shared by these experiments is described in Appendix Section D.2.1, and illustrated in Figure D.1.