Internet of Things (IoT) and machine learning are two important techniques in most industrial, business, agricultural, and medical applications. On the one hand, IoT systems keep producing massive sensory data as the input of various services. On the other hand, machine learning has obtained great success in vision, graphics, natural language processing, gaming, and controlling. This workshop calls for works demonstrating the most recent progress and contributions to learning in IoT. In particular, this workshop will focus on the follows (1) In-network federated learning, which does not need a center for sensory data sharing, but trains the machine learning model in a distributed fashion within the IoT; (2) Swarm learning that unites edge computing, blockchain-based peer-to-peer networking, without the need for a central coordinator. (3) Multi-agent reinforcement learning schemes for control of charging and moving, or decision making of communication, resource allocation, task scheduling, etc. This workshop especially encourages applications of learning techniques that make battery charging, event detection, localization in IoTs practical.
Chair: Tao Yu
| China Academy of Management Science, China
Tao Yu is a professor at China Academy of Management Science. His main research interests are passive positioning. More than 200 related academic papers have been published. An academic monograph "passive detection and positioning technology" funded by the state key publishing fund has been published by the national defense industry press. So far, three English monographs (coauthored) have been published. Some research results with great breakthrough significance are as follows: (1) Analytical analysis and potential application of double base path-differential positioning equation. Revealing the intrinsic correlation of passive localization equation. (2) Exploring the way to realize the phase difference position without phase fuzzy. (3) Coordinated location realizing by using external radiation sources and two aircraft in the case of unknown baseline. The classical geometrical relation is used to prove the innovative thinking.