Statistical Scalability for Streaming Data
Cambridge University
The profusion of sensor-based technologies in research and industrial systems is having a major influence on our ability to derive real-time insight and understanding about the world around us. However, this deluge of data streams also brings with it significant statistical challenge.In the past, there has been a tendency to focus purely on algorithmic scalability. However, increasingly we find new statistical challenges arising, requiring novel solutions. For example, the statistical analysis of data streams can involve computational and memory constraints due to a need to process data at source on limited hardware. An alternative challenge can arise with the problem of synthesizing information across multiple related streams, such as those observed in digital networks. Research is therefore needed to find more robust and effective methods for scaling data and, in particular, streaming data.This workshop is part of the six month programme at the Isaac Newton Institute (INI) on Sta