TechRxiv
performance_analysis_GAN_gen_techniques_IEEE_Access_Clean.pdf (3.66 MB)
Download file

Assessing Deep Generative Models on Time-Series Network Data

Download (3.66 MB)
preprint
posted on 21.04.2022, 14:11 by Muhammad Haris NaveedMuhammad Haris Naveed, Umair Hashmi, Nayab Tajved, Neha Sultan, Ali Imran
This paper explores whether Generative Adversarial Networks (GANs) can produce realistic network load data that can be utilized to train machine learning models in lieu of real data. In this regard, we evaluate the performance of three recent GAN architectures on the Telecom Italia data set across a set of qualitative and quantitative metrics. Our results show that GAN generated synthetic data is indeed similar to real data and forecasting models trained on this data achieve similar performance to those trained on real data.

Funding

1619346

1923669

NPRP12-S 0311-190302

History

Email Address of Submitting Author

umair.hashmi@seecs.edu.pk

ORCID of Submitting Author

0000-0001-8704-7132

Submitting Author's Institution

School of Electrical Engineering and Computer Science, National University of Sciences and Technology

Submitting Author's Country

Pakistan