GDPP: Learning Diverse Generations using Determinantal Point Processes
Authors: Mohamed Elfeki, Camille Couprie, Morgane Riviere, Mohamed Elhoseiny
ICML 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Our Generative DPP approach shows a consistent resistance to mode-collapse on a wide-variety of synthetic data and natural image datasets including MNIST, CIFAR10, and Celeb A, while outperforming stateof-the-art methods for data-efficiency, generation quality, and convergence-time whereas being 5.8x faster than its closest competitor. |
| Researcher Affiliation | Collaboration | 1University of Central Florida 2Facebook Artificial Intelligence Research 3King Abdullah University of Science and Technology. Correspondence to: Mohamed Elfeki <elfeki@cs.ucf.edu>, Couprie,Rivi ere <{coupriec,mriviere}@fb.com>, Mohamed Elhoseiny <mohamed.elhoseiny@kaust.edu.sa>. |
| Pseudocode | No | The paper does not contain any structured pseudocode or algorithm blocks. |
| Open Source Code | Yes | 1https://github.com/M-Elfeki/GDPP |
| Open Datasets | Yes | Stacked-MNIST A variant of MNIST (Le Cun, 1998)...; CIFAR-10 We evaluate the methods on CIFAR-10...; Celeb A Finally, to evaluate the performance of our loss on large-scale adversarial training... Celeb A dataset (Liu et al., 2018)... |
| Dataset Splits | No | The paper mentions training and testing on datasets but does not explicitly describe a validation set or specify train/validation/test splits. |
| Hardware Specification | No | The paper does not provide specific hardware details such as GPU or CPU models used for running the experiments. |
| Software Dependencies | No | The paper mentions using frameworks like DCGAN and VAE, but it does not specify version numbers for any software dependencies or libraries. |
| Experiment Setup | Yes | The same architecture is used for all methods and hyperparameters were tuned separately for each approach to achieve the best performance (See Appendix A for details). We train all models for 25K iterations...; We train all the models for 15,000 iterations...; We evaluate the methods on CIFAR-10 after training all the models for 100K iterations. |