Social Interaction in Trajectory Prediction with Memory Augmented Networks

Effective modeling of human interactions is of utmost importance when forecasting behaviors such as future trajectories. Each individual, with its motion, influences surrounding agents since everyone obeys to social non-written rules such as collision avoidance or group following. Prof. Alberto del Bimbo (Professor of Computer Engineering at the Department of Information Engineering and Director of MICC – Media Integration and Communication Center at the University of Florence, Italy) and his team believe that Memory Augmented Neural Networks are powerful models to effectively address this hard task. Prof. del Bimbo presents in his lecture a neural network based on an end-to-end trainable working memory, which acts as an external storage where information about each agent can be continuously written, updated and recalled. Memory Augmented Neural Networks have also the interesting capability of learning explainable cause-effect relationships between motions of different agents. Prof. del Bimbo shows state-of-the-art results on multiple trajectory forecasting datasets.

Tuesday, 07-05-2022, 05:00 p.m. – 06:00 p.m., virtual event

Further details:

This is an event of I-AIDA – International Artificial Intelligence Doctoral Academy.