1. This article proposes a novel digital twin (DT) empowered IIoT (DTEI) architecture, which captures the properties of industrial devices for real-time processing and intelligent decision making.
2. To optimize federated learning (FL) to construct the DTEI model, a deep reinforcement learning method is developed for the selection process of IIoT devices in FL.
3. An asynchronous FL scheme is proposed to address the discrete effects caused by heterogeneous IIoT devices.
The article provides an overview of the development of Industrial Internet of Things (IIoT), digital twin (DT), and federated learning (FL). It presents a novel digital twin empowered IIoT architecture and proposes a deep reinforcement learning method for selecting IIoT devices with high utility values in FL, as well as an asynchronous FL scheme to address the discrete effects caused by heterogeneous IIoT devices. The article is written in an objective manner and provides evidence from experiments to support its claims. However, it does not explore any potential risks associated with using this technology or discuss any counterarguments that may exist. Additionally, it does not provide any information on how this technology could be used in practice or what implications it may have on society or industry. Furthermore, there is no discussion about possible biases or sources of bias that could affect the results presented in the article. In conclusion, while this article provides an interesting overview of current research into optimizing federated learning with deep reinforcement learning for digital twin empowered industrial IoT, more research needs to be done to explore potential risks and implications before this technology can be implemented in practice.