Multiple-Instance Learning of Real-Valued Data

By Daniel R. Dooly, Qi Zhang, Sally A. Goldman, Robert A. Amar. 2001

162 citations recorded by the source corpus.

Subfields: Computational Theory and Mathematics, Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computational Theory and Mathematics

In Computer Vision and Pattern Recognition

Citations in the corpus

Data-efficient weakly supervised learning for low-resource audio event detection using deep learning (2018, background): “As indicated in [3], the value of a bag is fully determined by its instance with the maximal output, regardless how many real positive or negative instances in the bag.”
Multi-instance learning by treating instances as non-I.I.D. samples (2008, methodology): “Chen and Wang [7] proposed the DD-SVM method which employed Diverse Density [16] to learn a set of instance prototypes and then maps the bags to a feature space based on the instance pro-totypes.”
Fast Bundle Algorithm for Multiple-Instance Learning (2012, background): “Index Terms—Artificial intelligence, machine learning, nonsmooth optimization, bundle methods, multiple-instance learning, ranking, medicine and science.”
EM-DD: An Improved Multiple-Instance Learning Technique (2001, methodology): “In this paper , we present a general-purpose MI learning technique (EM-DD) that combines EM [3] with the extended DD [1] algorithm.”
Maximum Margin Multiple Instance Clustering With Applications to Image and Text Clustering (2011, background): “, the maximal, minimal and the average Hausdorff distance [1][30][33].”
Multi-Instance Learning : A Survey (2004, background): “…algorithm is almost the same as that of the classical BP algorithm, it is evident that the key of BP-MIP lies in the formal definition of the multi-instance error function 1 , which is the function used to measure the error of the neural network and therefore is the objective to be optimized.”
Multi-instance clustering with applications to multi-instance prediction (2009, methodology): “Both maximal and minimal Hausdorfi distances have been successfully applied to the StdMIP prob- lems [17] and MIR problems [ 4 ].”
Multi-instance multi-label learning (2008, background): “Later, multi-instance regression with real-valued outputs was studied [2,52], and different versions of generalized multi-instance learning have been defined [58, 68].”
Multiple instance classification: Review, taxonomy and comparative study (2013, methodology): “As analyzed in [10], the types of solution s can be categorized in a similar manner to the one employed for MIC (including instance-leve l solutions using the standard MI assumption [38], embedded-space solutions [35] and bag-sp ce solutions [39]).”
Two-stage sampled learning theory on distributions (2014, background): “On the machine learning side, multiple instance learning [2, 3, 4] can be thought of in this way, in the case where each instance in a labeled bag is an i.”
Multi-Instance Kernels (2002, methodology): “For example, by simply plugging our kernel into SVM regression, the only very recently formulated problem of MI regression can be tackled (Amar et al., 2001; Ray & Page, 2001).”
Marginalized Multi-Instance Kernels (2007, unknown): “In this real-valued setting, we follow the extended DD model in [Amar et al., 2001], and have: P (cij |xij) = ( 1 − |cij − e−‖h−xij‖ 2 | ) /Z(xij), (12) where Z(xij) ensures that ∫ 1 0 P (cij |xij)dcij = 1.”
Multiple-instance discriminant analysis (2014, background): “[42] when they were studying to generate artificial datasets to evaluate the classification performance of multiple-instance learners.”