Proposal Description

Abstract: Time Series Analysis is widely applied and remains popular owing to its applicability in a plethora of fields. With the field emerging, advances in Machine Learning made their way into Time Series Analysis. Machine Learning powered and/or improved models showed potential to improved forecasting as well as the interpretability of complex systems. Resulting models benefit from the established nature of Time Series Models and the innovative nature of Machine Learning techniques. The kind of systems range from sparse to dense, biology to finance, univariate to multivariate, you name it. In this session, the talks will discuss methodological advances stemming from the field of machine learning and time series analysis for modelling, monitoring, analyzing, and conducting inference within systems of a diverse nature.

Justification: Time series analysis is an important field, complemented by the emerging field of Machine Learning, where many challenges remain. Important challenges are the modelling, analysis, and forecasting of large dynamic systems. Methodologies are applicable in a wide variety of fields such as finance, biology, macroeconomy, and even social networks. At the same time, recent advances from machine learning proved their efficacy in improving and complementing time series models. This session aims to showcase new advances in solving these important challenges and aims to give researchers a platform to share their recent work in this area, creating an active discussion platform for new advances.