Abstract:Multiple instance learning (MIL) is a form of weakly supervised learning where training instances are arranged in sets, called bags, and a label is provided for the entire bag. This formulation is gaining interest because it naturally fits various problems and allows to leverage weakly labeled data. Consequently, it has been used in diverse application fields such as computer vision and document classification. However, learning from bags raises important challenges that are unique to MIL. This paper provides a comprehensive survey of the characteristics which define and differentiate the types of MIL problems. Until now, these problem characteristics have not been formally identified and described. As a result, the variations in performance of MIL algorithms from one data set to another are difficult to explain. In this paper, MIL problem characteristics are grouped into four broad categories: the composition of the bags, the types of data distribution, the ambiguity of instance labels, and the task to be performed. Methods specialized to address each category are reviewed. Then, the extent to which these characteristics manifest themselves in key MIL application areas are described. Finally, experiments are conducted to compare the performance of 16 state-of-the-art MIL methods on selected problem characteristics. This paper provides insight on how the problem characteristics affect MIL algorithms, recommendations for future benchmarking and promising avenues for research. Code is available on-line at https://github.com/macarbonneau/MILSurvey.

Diversified Dictionaries for Multi-Instance Learning

Multiple-Instance Learning from Pairwise Comparison Bags

Multi-Instance Learning with Discriminative Bag Mapping

Mild: Multiple-Instance Learning Via Disambiguation

Max-Margin Multiple-Instance Dictionary Learning.

Multiple Instance Learning for Unilateral Data

Multiple instance learning with bag dissimilarities

Salient instance selection for multiple-instance learning

Learning Multi-Instance Deep Discriminative Patterns for Image Classification.

Image Annotation by Multiple-Instance Learning with Discriminative Feature Mapping and Selection

Multiple instance learning from similarity-confidence bags

Improving Multiple-Instance Learning via Disambiguation by Considering Generalization.

Deep multiple instance selection

Multiple-Instance Learning from Similar and Dissimilar Bags

Multiple-instance Learning with Pairwise Instance Similarity

Multiple-Instance Learning from Unlabeled Bags with Pairwise Similarity

Bag-level active multi-instance learning

Learning Instance Specific Distance for Multi-Instance Classification.

Multiple instance learning based on positive instance selection and bag structure construction

Multiple instance learning: A survey of problem characteristics and applications

Bag dissimilarity regularized multi-instance learning