Decentralized Attribution of Generative Models

Authors: Changhoon Kim, Yi Ren, Yezhou Yang

ICLR 2021 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Our method is validated on MNIST, Celeb A, and FFHQ datasets.
Researcher Affiliation Academia Changhoon Kim1, , Yi Ren2, , Yezhou Yang1 School of Computing, Informatics, and Decision Systems Engineering1 School for Engineering of Matter, Transport, and Energy2 Arizona State University {kch, yiren, yz.yang}@asu.edu
Pseudocode Yes Algorithm 1: Training of Gφ input : φ, G0 output: Gφ, γ
Open Source Code Yes 1https://github.com/ASU-Active-Perception-Group/decentralized_ attribution_of_generative_models
Open Datasets Yes We validate these rules using DCGAN (Radford et al., 2015) and Style GAN (Karras et al., 2019a) on benchmark datasets including MNIST (Le Cun & Cortes, 2010), Celeb A (Liu et al., 2015), and FFHQ (Karras et al., 2019a).
Dataset Splits No The paper uses the term 'validation' in the context of validating theorems, not as a dataset split (e.g., train/validation/test percentage or count).
Hardware Specification Yes All experiments are conducted on V100 Tesla GPUs.
Software Dependencies No The paper mentions software components and frameworks like PyTorch (inferred from Kornia citation) and specific implementations for blurring and JPEG conversion, but it does not provide specific version numbers for these software dependencies.
Experiment Setup Yes We adopt the Adam optimizer for training. Training hyper-parameters are summarized in Table 3.