Multi-View Disentangled Representation
Learning effective representations for data with multiple views is crucial in machine learning and pattern recognition. Recently great efforts have focused on learning unified or latent representations to integrate information from different views for specific tasks. These approaches generally assume simple or implicit relationships between different views and as a result are not able to flexibly and explicitly depict the correlations among these views. To address this, we firstly propose the definition and conditions for multi-view disentanglement providing general instructions for disentangling representations between different views. Furthermore, a novel objective function is derived to explicitly disentangle the multi-view data into a shared part across different views and a (private) exclusive part within each view. Experiments on a variety of multi-modal datasets demonstrate that our objective can effectively disentangle information from different views while satisfying the disentangling conditions.
PDF Abstract
CelebA
