Machine learning has seen wild success in solving problems from various domains, ranging from classical examples such as computer vision, natural language processing, voice recognition, to recent adventures such as self-driving cars and strategic game playing (Go).

We believe machine learning has strong potential in solving problems in computer networks. Networks are becoming more and more complicated with the growing demand for cloud computing and big data. A production network usually involves a multitude of devices, runs a multitude of protocols, and supports a multitude of applications. Traditional approaches of designing, deploying, and managing protocols face significant challenges in these complex networks. Machine learning represents a different and potentially rewarding approach to solving these challenges: its data-driven nature allows it to intelligently learn the complicated network environment and dynamically adjust protocols, with little manual effort.

Research on machine learning in networks is still at an early stage. There is in general a lack of venue dedicated for discussion, promotion, and dissemination of research on machine learning in computer networks. NetworkML aims to bring together researchers and practitioners in computer networks, systems, and machine learning to engage in a lively debate on the theory and practice of using machine learning in computer networking research.