A Cyber Physical System Crowdsourcing Inference Method Based on Tempering: An Advancement in Artificial Intelligence Algorithms

Activity selection is critical for the smart environment and Cyber-Physical Systems (CPSs) that can provide timely intelligent services, especially as number of connected devices increasing at an unprecedented speed. As it important to collect labels by various agents in CPSs, crowdsourcing inference algorithms are designed help acquire accurate involve high-level knowledge. However, there some limitations algorithm existing literature such incurring extra budget algorithms, inability scale appropriately, requiring knowledge prior distribution, difficulties implement these or generating local optima. In this paper, we a method with variational tempering obtains ground truth well considers both reliability workers difficulty level tasks ensure optimum. The numerical experiments real-world data indicate our novel performs better than advancing algorithms. Therefore, paper provides new efficient CPSs machine learning, thus, makes contribution literature.