Pietro Mascolo: Timeseries on steroids with RETAIN networks

Timeseries on steroids with RETAIN networks Pietro Mascolo Sequence modeling plays a crucial role in machine learning applications, from time series forecasting to natural language processing and healthcare analytics. This talk introduces RETAIN (Reversed Time Attention Model), a powerful architecture designed for interpretable sequence modeling in healthcare but adaptable to other domains. We’ll explore how RETAIN leverages attention mechanisms in a reversed time order to prioritize significant events in sequences, ensuring both predictive performance and interpretability. Using Python and libraries like PyTorch and Scikit-learn, this session will demonstrate building and applying RETAIN models, discussing their architecture, practical applications, and potential customizations for diverse use cases. Talk room 2 (200 people) Sat 11:55 am - 12:25 pm