Jin Wang is an Enrichment student at The Alan Turing Institute. He is also a Ph.D. student at the University of Exeter. He received both B.Eng. and M.Eng. degree in Computer System Architecture from the University of Electronic Science and Technology of China (UESTC), Chengdu, China.
Jin Wang's main research interest is deep learning (DL) based methods for distributed system optimization and maintenance. Especially, he is adapting deep reinforcement learning-based methods for the task offloading, service migration problems in multi-access edge computing (MEC). He also handles the anomalous time-series data generation problems by adapting Generative Adversarial Networks (GAN) and Variational Autoencoder (VAE) in Cloud platforms. He has published several research papers within these areas in prestigious international journals and magazines (e.g., IEEE Journal of Internet of things, IEEE Communications Magazine) and at reputable international conferences (e.g., IEEE Globecom).
He focuses on applying DL-based methods to solve the optimization and maintenance problems in cloud and MEC systems. Effective optimization and maintenance for these systems are vitally important because they can help significantly improve the performance, reliability, and resource utilization rate. DL can be an enabler to achieve intelligent maintenance and optimization on these complex systems. Especially, to improve the systems’ performance and save energy, he adapted (meta) reinforcement learning-based methods to achieve smarter decision-making for task offloading and server migration problems in MEC, considering user mobility. In addition, to synthesize the training data for network fault prediction and detection in Cloud platforms, he introduced deep generative models (i.e., variants of GAN and VAE) to solve the anomalous time-series data generation problem.