AI for Network Sciences: From understanding data to reliable deployment in wireless communications
The Technical Doctoral School of IT and Design at Aalborg University
Welcome to AI for Network Sciences: From understanding data to reliable deployment in wireless communications
Description: Modern AI applications face novel challenges, particularly in network sciences, where ML algorithms must handle data from diverse sources, respect data privacy constraints, and optimize the learning process across multiple agents. Offering valid statistical conclusions by learning from unreliable data is critical for several real-world applications. However, it may be challenging to dive into AI for network sciences, as it requires understanding many concepts from probability, statistics, distributed AI/ML, and data markets. This course addresses those challenges by covering: (i) the fundamentals of network sciences and the types of data encountered in this domain, (ii) collaborative and distributed machine learning algorithms, and (iii) reliable guarantees based on conformal prediction and decision theory.