Improving Content-Invariance in Gated Autoencoders for 2D and 3D Object Rotation
This addresses a specific issue in relation learning tasks for computer vision, but it is incremental as it builds on existing GAE methods.
The paper tackled the problem of improving content-invariance in mapping codes for 2D and 3D rotated objects in Gated Autoencoders, achieving substantial improvement by adding a regularization term for symmetric cross-reconstruction error.
Content-invariance in mapping codes learned by GAEs is a useful feature for various relation learning tasks. In this paper we show that the content-invariance of mapping codes for images of 2D and 3D rotated objects can be substantially improved by extending the standard GAE loss (symmetric reconstruction error) with a regularization term that penalizes the symmetric cross-reconstruction error. This error term involves reconstruction of pairs with mapping codes obtained from other pairs exhibiting similar transformations. Although this would principally require knowledge of the transformations exhibited by training pairs, our experiments show that a bootstrapping approach can sidestep this issue, and that the regularization term can effectively be used in an unsupervised setting.