GAN Memory with No Forgetting

Authors: Yulai Cong, Miaoyun Zhao, Jianqiao Li, Sijia Wang, Lawrence Carin

NeurIPS 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experiments demonstrate the superiority of our method over existing approaches and its effectiveness in alleviating catastrophic forgetting for lifelong classification problems.
Researcher Affiliation Academia Yulai Cong Miaoyun Zhao Jianqiao Li Sijia Wang Lawrence Carin Department of Electrical and Computer Engineering Duke University
Pseudocode No The paper does not contain structured pseudocode or algorithm blocks.
Open Source Code Yes Code is available at https:// github.com/Miaoyun Zhao/GANmemory_Lifelong Learning.
Open Datasets Yes To demonstrate the superiority of our GAN memory over existing replay-based methods, we design a challenging lifelong generation problem consisting of 6 perceptually-distant tasks/datasets (see Figure 5): Flowers [57], Cathedrals [99], Cats [97], Brain-tumor images [15], Chest X-rays [35], and Anime faces.10 The GP-GAN [49] trained on the Celeb A [43] (D0) is selected as the base; other well-behaved GAN models may readily be considered.
Dataset Splits No The paper mentions training data and testing performance but does not specify explicit train/validation/test dataset splits or their sizes, or reference a standard split that includes a validation set for reproducibility.
Hardware Specification Yes The Titan Xp GPU used was donated by the NVIDIA Corporation.
Software Dependencies No The paper does not provide specific software dependencies with version numbers (e.g., Python, PyTorch, TensorFlow versions) that would allow for reproducible setup of the environment.
Experiment Setup No Detailed experimental settings are given in Appendix A.