We suggest a novel approach for classification that extracts the most out of the two simple yet defining features of a flow: packet sizes and inter-arrival times. We employ a model that uses the inter-arrival times to parameterize the derivative of the flow hidden-state using a neural network (Neural ODE). We compare our results with a solution that uses the same data without the ODE solver and show the benefit of this approach. 
We introduce a novel approach for Autonomous System (AS) embedding using deep learning based on only BGP announcments. Using these vectors we able to solve multiple important classification problem such as AS business types, AS Types of Relationship (ToR) and even IP hijack detection. 
We introduce a novel approach for encrypted Internet traffic classification and application identification by transforming basic flow data into a picture, a FlowPic, and then using known image classification deep learning techniques, Convolutional Neural Networks (CNNs), to identify the flow category (browsing, chat, video, etc.) and the application in use. 
We introduce a novel deep learning layer, called Source-Aware Self-Attention (SASA), which is an extension of the attention mechanism. SASA learns each data source’s confidence and combines this score with the attention of each router in the route to point out the most problematic one. 