Dictionary-based multi-instance learning method with universum information

TL;DR: A Bayesian approach to supervised learning, which leads to sparse solutions; that is, in which irrelevant parameters are automatically set exactly to zero, and involves no tuning or adjustment of sparseness-controlling hyperparameters.

TL;DR: A discriminative dictionary learning algorithm, called the locality-constrained and label embedding dictionary learning (LCLE-DL) algorithm, was proposed for image classification, which can achieve better performance than some state-of-the-art algorithms.

TL;DR: This paper proposes a multimodal task-driven dictionary learning algorithm under the joint sparsity constraint (prior) to enforce collaborations among multiple homogeneous/heterogeneous sources of information and presents an extension of the proposed formulation using a mixed joint and independent sparsity prior, which facilitates more flexible fusion of the modalities at feature level.

TL;DR: A general bag dissimilarities framework for multiple instance learning is explored and several alternatives to define a dissimilarity between bags are shown and discussed, which definitions are more suitable for particular MIL problems.