Deep Generative Models with Learnable Knowledge Constraints
Authors: Zhiting Hu, Zichao Yang, Russ R. Salakhutdinov, LIANHUI Qin, Xiaodan Liang, Haoye Dong, Eric P. Xing
NeurIPS 2018 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Experiments on human image generation and templated sentence generation show models with learned knowledge constraints by our algorithm greatly improve over base generative models. |
| Researcher Affiliation | Collaboration | Zhiting Hu, Zichao Yang, Ruslan Salakhutdinov, Xiaodan Liang, Lianhui Qin, Haoye Dong, Eric P. Xing Carnegie Mellon University, Petuum Inc. {zhitingh,zichaoy,rsalakhu,xiaodan1}@cs.cmu.edu, eric.xing@petuum.com |
| Pseudocode | Yes | Algorithm 1 Joint Learning of Deep Generative Model and Constraints |
| Open Source Code | No | The paper does not explicitly state that source code for the described methodology is provided or made publicly available, nor does it include a link to a code repository. |
| Open Datasets | Yes | We follow the previous work [37] and obtain from Deep Fashion [35] a set of triples (source image, pose keypoints, target image) as supervision data. and Paired (template, sentence) data is obtained by randomly masking out different parts of sentences from the IMDB corpus [8]. |
| Dataset Splits | No | The paper mentions using a 'test set' and 'test cases' but does not specify the train/validation/test dataset splits (e.g., percentages or exact counts) needed for reproduction. |
| Hardware Specification | No | The paper does not provide specific hardware details (e.g., GPU/CPU models, memory specifications) used for running its experiments. |
| Software Dependencies | No | The paper does not provide specific ancillary software details with version numbers (e.g., library or solver names with version numbers) needed to replicate the experiment. |
| Experiment Setup | No | The paper describes the general setup for the base models and constraints (e.g., 'residual block architecture', 'L1 distance loss') but does not provide specific numerical hyperparameters like learning rates, batch sizes, number of epochs, or optimizer settings needed for full reproducibility of the experimental setup. |