We present sparse filtering, a simple new algorithm which is efficient and only has one hyperparameter, the number of features to learn.

Unsupervised feature learning has been shown to be effective at learning repre- sentations that perform well on image, video and audio classification.

The sparsefilt function creates a nonlinear transformation of input features to output features. The transformation is based on optimizing an objective ...

In this paper we present a theoretical analysis to understand sparse filtering, a recent and effective algorithm for unsupervised learning.

SparseFiltering uses sparse filtering to learn a transformation that maps input predictors to new predictors. Creation. Create a SparseFiltering object ...

May 27, 2015 · SF is an algorithm or learning module to perform unsupervised feature distribution learning that generates sparse representations. Given a ...

Evaluated sparse filtering features on natural images, image classification (STL Dataset), audio classification (TIMIT). Results comparable to state-of-the-art ...

In this paper we present a theoretical analysis to understand sparse filtering, a recent and effective algorithm for unsupervised learning.

Unsupervised feature learning based on sparse-filtering. License. View license · 55 stars 13 forks Branches Tags Activity.

Directional Sparse FIltering introduced by Nguyen et al. in Learning complex-valued latent filters with absolute cosine similarity.