The Principal Component Analysis (PCA) is considered to be a quintessential data preprocessing tool in many machine learning applications. But the high dimensionality and massive scale of data in several of these applications means the traditional centralized PCA solutions are fast becoming irrelevant for them. 

The Principal Component Analysis (PCA) is considered to be a quintessential data preprocessing tool in many machine learning applications. But the high dimensionality and massive scale of data in several of these applications means the traditional centralized PCA solutions are fast becoming irrelevant for them. 

In past decades, conventional communication primarily focused on how to accurately and effectively transmit symbols, which is categorized as the first level of communications by Shannon and Weaver. With the developments of cellular communication systems, the achieved transmission rate is gradually approaching to the Shannon limit. 

As the standardization of 5G gradually solidifies, researchers are speculating what 6G will be. One common theme is that radio sensing functionality would be integrated into 6G networks in a low-cost and fast manner. 

Over the past fifteen years, “cognition” has emerged as an enabling technology for incorporating learning and adaptivity on both transmit and receive to optimize or make more robust the radar performance in dynamic environments.The term ‘cognitive radar’ was introduced for the first time by Dr. Simon Haykin in 2006, but the foundations of the cognitive systems date back several decades to research on knowledge-aided signal processing, and adaptive radar design.

Substantial progress has been made recently on developing provably accurate and efficient algorithms for low-rank matrix factorization via nonconvex optimization. While conventional wisdom often takes a dim view of nonconvex optimization algorithms due to their susceptibility to spurious local minima, simple iterative methods such as gradient descent have been remarkably successful in practice. 

This webinar, part of the new IEEE JSTSP webinar series on recent special issues (Sis), will overview the Joint Communication and Radar Sensing (JCR) for Emerging Applications. The webinar will start with a brief motivation of the area of JCR, followed by a summary of the technical papers that appear in the SI.

In this talk, we investigate the model-driven deep learning for multiple input-multiple output (MIMO) detection. In particular, the MIMO detector is specially designed by unfolding an iterative algorithm and adding some trainable parameters. 

In this talk, we discuss a new transform technique for solving fractional programming (FP), i.e., a family of optimization problems with ratio terms. The classic FP techniques such as the Charnes-Cooper and Dinkelbach’s methods typically deal with a single ratio, and in general, do not work for multiple ratios. 

Many important application domains generate distributed collections of heterogeneous local datasets. These local datasets are related via an intrinsic network structure that arises from domain-specific notions of similarity between local datasets. Networked federated learning aims at learning a tailored local model for each local dataset.