Solving the Multiple Instance Problem with Axis-Parallel Rectangles

By Thomas G. Dietterich, R. Lathrop, Tomas Lozano-Perez. 1997

How researchers see this paper

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

In Statistical Learning Theory

Citations in the corpus

End-to-End Learning of Visual Representations From Uncurated Instructional Videos (2019, methodology): “In this work, we present MIL-NCE, a new approach marrying the noise contrastive estimation (NCE) framework [29] with multiple instance learning [20].”
Objects that Sound (2017, methodology): “To operationalize this, we formulate the problem in the Multiple Instance Learning (MIL) framework [9].”
Set Transformer (2018, background): “Multiple instance learning (Dietterich et al., 1997; Maron & Lozano-P´erez, 1998) is an example of such a set-input problem, where a set of instances is given as an input and the corresponding target is a label for the entire set.”
Learning Theory for Distribution Regression (2014, background): “Multi-instance learning: An alternative paradigm in learning when the inputs are “bags of objects” is to simply treat each input as a finite set ; this leads to the multi-instance learning task (MIL, see Dietterich et al., 1997; Ray and Page, 2001; Dooly et al., 2002).”
Distribution Kernel Methods for Multiple-Instance Learning (2013, methodology): “The multiple-instance (MI) learning (MIL) framework is an extension of supervised learning motivated by the above problem (Dietterich, Lathrop, and Lozano-Pérez 1997), and encodes this relationship between an observed label and a set instances responsible for that label.”
Pushing the Boundaries of Boundary Detection using Deep Learning (2015, methodology): “Our approach builds on Kokkinos (2010a), where Multiple Instance Learning (MIL) (Dietterich et al., 1997) is used to accommodate orientation inconsistencies during the learning of an orientation-sensitive boundary detector.”
Attention-based Deep Multiple Instance Learning (2018, neutral): “This scenario is called multiple instance learning (MIL) (Dietterich et al., 1997; Maron & Lozano-P´erez, 1998) or, learning from weakly annotated data (Oquab et al., 2014).”
Multiple instance learning for sparse positive bags (2007, methodology): “A variety of MIL algorithms have been developed over the past ten years (Dietterich et al., 1997; Gartner et al., 2002; Andrews et al., 2003); however, the sim- Appearing in Proceedings of the 24 th International Conference on Machine Learning, Corvallis, OR, 2007.”
Deep Multi-instance Networks with Sparse Label Assignment for Whole Mammogram Classification (2016, background): “The bag is negative if and only if all instances are negative [8].”
Machine Learning Methods for Histopathological Image Analysis (2017, background): “This setting is a problem of multiple instance learning [51,108] or weakly-supervised learning [50,52].”
Visual tracking with online Multiple Instance Learning (2009, methodology): “Numerous algorithms have been proposed for solving the MIL problem [9, 2, 23].”
DTFD-MIL: Double-Tier Feature Distillation Multiple Instance Learning for Histopathology Whole Slide Image Classification (2022, background): “A large proportion of existing weakly supervised works for WSI classiﬁcation are characterized as “multiple instance learning” (MIL) [1,5,8,25].”
Learning Neural Networks with Two Nonlinear Layers in Polynomial Time (2017, background): “nstance Learning Multiple Instance Learning (MIL) is a generalization of supervised classiﬁcation in which a label is assigned to a bag, that is, a set of instances, instead of an individual instance [DLLP97]. The bag label is induced by the labels of the instances in it. The goal we focus on in thi”
Multiple-Instance Learning of Real-Valued Data (2001, methodology): “The first empirical study of Dietterich et al. (1997) used real data for the problem of predicting whether or not a synthetic molecule binds to the musk receptor.”