Continual Learning Systems for a Data Intensive World

The February 1, 2017 DataBytes webinar, “Continual Learning Systems for a Data-Intensive World,” was presented by Gail Rosen, associate professor in the Department of Electrical and Computer Engineering at Drexel University.

About the Talk: Real-World data is often generated continuously and is non-stationary. Therefore, as we develop and implement new technology to collect new data, new information can be incompatible with the old datasets due to the change of dimensionality and knowledge. Hence, incremental and active learning systems can address these issues. They can incrementally update the existing decision boundaries based on new data. We will present advances in incremental learning and unsupervised and semi-supervised classification, and their applications to citation and genomic data.