The Power of Variational and Hybrid Multi-task Models for Image Analysis

OVERVIEW

The conceptual philosophy of joint models (aka multi-tasking) for image analysis has demonstrated powerful results especially for complex data including medical. This improved performance is due to the fact that - by sharing representation between tasks and carefully intertwining them, one can create synergies across challenging problems and reduce error propagation. This results in boosting the accuracy of the outcomes whilst achieving better generalisation capabilities than sequential models and keeping reasonable computational cost. These advantages have motivated a fast development of new algorithmic approaches including hybrid techniques (variational + machine learning techniques/model-based + data-driven methods). In this session, we aim to discuss the recent developments on joint models ranging from pure variational to hybrid approaches.