Contractive Autoencoder
Description
A Contractive Autoencoder is an autoencoder that adds a penalty term to the classical reconstruction cost function. This penalty term corresponds to the Frobenius norm of the Jacobian matrix of the encoder activations with respect to the input. This penalty term results in a localized space contraction which in turn yields robust features on the activation layer. The penalty helps to carve a representation that better captures the local directions of variation dictated by the data, corresponding to a lower-dimensional non-linear manifold, while being more invariant to the vast majority of directions orthogonal to the manifold.
Papers Using This Method
Scalable Self-Supervised Representation Learning from Spatiotemporal Motion Trajectories for Multimodal Computer Vision2022-10-07Reachability Embeddings: Scalable Self-Supervised Representation Learning from Mobility Trajectories for Multimodal Geospatial Computer Vision2021-10-24Predicting Sample Collision with Neural Networks2020-06-30Majorization Minimization Technique for Optimally Solving Deep Dictionary Learning2019-12-11Towards Deep Neural Network Architectures Robust to Adversarial Examples2014-12-11