Publications News

The importance of developing cognitive computation and cognitive systems has long been of interest. This research field is now rapidly expanding with the current influx of artificial intelligence innovations. Nowadays, machine learning solutions are dramatically transforming the lives of humans and are having a profound impact on a wide range of disciplines and industries. However, machine learning is not omnipotent, especially under the circumstance of cognitive systems operating in open and dynamic environments.

What Should We Learn? Uncertain Multicriteria Decision Making Using Evolutionary Algorithms

Uncertain multicriteria decision making is to select or rank objects based on the evaluation done by the decisionmaker on several criteria under uncertainty. Uncertain multicriteria decision making has been proved as a useful means in diverse fields such as management, finance, economics, education, environmental protection, medicine, engineering, and so on. Due to numerous successful applications, it becomes more and more prevailing.

Going Toward 6G

20190510LiY-SPM-going 2 6G.png

SPS Publications Alert: May 2019 Tables of Contents

TOC_general.jpg

SPS Publications Alert: April 2019 Tables of Contents

TOC_general.jpg

The list of tables of contents (TOCs) are now available for April 2019. The TOCs are formatted to the style of its publication and offers not only links to the full issue on the front cover, but also links to the individual articles within each TOC to take you to IEEEXplore®.

What Should We Learn? Special Issue on Computational Intelligence in Data-Driven Optimization

0319_spm_A1.jpg

Most evolutionary algorithms and other meta-heuristic search methods typically assume that there are explicit objective functions available for fitness evaluations. In the real world, such explicit objective functions may not exist in many cases. Instead, computationally very intensive numerical simulations, such as computational fluid dynamic simulations or finite element analysis, must be conducted.

What Should We Learn? Learning ReLU Networks on Linearly Separable Data

General_SP.jpg

Neural networks with rectified linear unit (ReLU) activation functions (a.k.a. ReLU networks) have achieved great empirical success in various domains. Nonetheless, existing results for learning ReLU networks either pose assumptions on the underlying data distribution being, e.g., Gaussian, or require the network size and/or training size to be sufficiently large.

Special Issue on Adaptive and Scalable Communication Networks

general_iStock_000013857769_Small.jpg

In the special issue in Proceedings of the IEEE in April 2019, the editors have collected and presented recent works on innovative approaches and emerged technologies for coping with dynamicity, heterogeneity, and the scale, which have been central to (or even enablers of) recent advances in communications and networking technologies.

SPS Publications Alert: March 2019 Tables of Contents

March 2019