Compressing Image-to-Image Translation GANs Using Local Density Structures on Their Learned Manifold
Authors: Alireza Ganjdanesh, Shangqian Gao, Hirad Alipanah, Heng Huang
AAAI 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Our experiments on image translation GAN models, Pix2Pix and Cycle GAN, with various benchmark datasets and architectures demonstrate our method s effectiveness. |
| Researcher Affiliation | Academia | 1 Department of Computer Science, University of Maryland College Park, College Park, MD 20742, USA 2 Department of Electrical and Computer Engineering, University of Pittsburgh, Pittsburgh, PA 15261, USA 3 Department of Mechanical Engineering and Materials Science, University of Pittsburgh, Pittsburgh, PA 15261, USA |
| Pseudocode | Yes | We provide our pruning algorithm and details of the calculation of o and h in the supplementary. |
| Open Source Code | Yes | We provide our pruning algorithm and details of the calculation of o and h in the supplementary. |
| Open Datasets | Yes | Our experiments on image translation GAN models, Pix2Pix and Cycle GAN, with various benchmark datasets and architectures demonstrate our method s effectiveness. |
| Dataset Splits | No | The paper refers to "training dataset" and "training samples" but does not specify explicit splits like percentages or sample counts for train/validation/test. |
| Hardware Specification | No | The paper does not provide specific hardware details (e.g., GPU/CPU models, memory) 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 | Yes | For all experiments, we set λ1 = 3.0, λ2 = 0.1, and the number of neighbor samples k = 5 during pruning. We also find that our model is not very sensitive to the choice of λ1 and λ2. (more details in ablation experiments) We set the number of pruning epochs to 10% of the original model s training epochs. |