artificial intelligence for students
14.5K views | +0 today
Follow
 
Scooped by Scott Turner
onto artificial intelligence for students
April 15, 2020 10:35 AM
Scoop.it!

Predictability of temporal networks quantified by an entropy-rate-based framework

Predictability of temporal networks quantified by an entropy-rate-based framework | artificial intelligence for students | Scoop.it
Networks or graphs are mathematical descriptions of the internal structure between components in a complex system, such as connections between neurons, interactions between proteins, contacts between individuals in a crowd, and interactions between users in online social platforms. The links in most real networks change over time, and such networks are often called temporal networks. The temporality of links encodes the ordering and causality of interactions between nodes and has a profound effect on neural network function, disease propagation, information aggregation and recommendation, emergence of cooperative behavior, and network controllability. Increasing research has focused on mining the patterns in a temporal network and predicting its future evolution using machine learning techniques, especially graph neural networks. However, how to quantify the predictability limit of a temporal network, i.e. the limit that no algorithm can go beyond, is still an open question.
No comment yet.
artificial intelligence for students
artificial intelligence for students - postgraduate and undergraduate
Curated by Scott Turner