Dissimilarity-based ensembles for multiple instance learning

V. Cheplygina, D.M.J. Tax, M. Loog

Publikation: Artikel i tidsskrift og konference artikel i tidsskriftTidsskriftartikelForskningpeer review


In multiple instance learning, objects are sets (bags) of feature vectors (instances) rather than individual feature vectors. In this paper, we address the problem of how these bags can best be represented. Two standard approaches are to use (dis)similarities between bags and prototype bags, or between bags and prototype instances. The first approach results in a relatively low-dimensional representation, determined by the number of training bags, whereas the second approach results in a relatively high-dimensional representation, determined by the total number of instances in the training set. However, an advantage of the latter representation is that the informativeness of the prototype instances can be inferred. In this paper, a third, intermediate approach is proposed, which links the two approaches and combines their strengths. Our classifier is inspired by a random subspace ensemble, and considers subspaces of the dissimilarity space, defined by subsets of instances, as prototypes. We provide insight into the structure of some popular multiple instance problems and show state-of-the-art performances on these data sets.
TidsskriftIEEE Transactions on Neural Networks and Learning Systems
Udgave nummer6
Sider (fra-til)1379-1391
Antal sider13
StatusUdgivet - jun. 2016
Udgivet eksterntJa


Dyk ned i forskningsemnerne om 'Dissimilarity-based ensembles for multiple instance learning'. Sammen danner de et unikt fingeraftryk.