MIS-Boost: Multiple Instance Selection Boosting

09/12/2011
by   Emre Akbaş, et al.
0

In this paper, we present a new multiple instance learning (MIL) method, called MIS-Boost, which learns discriminative instance prototypes by explicit instance selection in a boosting framework. Unlike previous instance selection based MIL methods, we do not restrict the prototypes to a discrete set of training instances but allow them to take arbitrary values in the instance feature space. We also do not restrict the total number of prototypes and the number of selected-instances per bag; these quantities are completely data-driven. We show that MIS-Boost outperforms state-of-the-art MIL methods on a number of benchmark datasets. We also apply MIS-Boost to large-scale image classification, where we show that the automatically selected prototypes map to visually meaningful image regions.

READ FULL TEXT

page 7

page 8

page 9

research
06/17/2022

DGMIL: Distribution Guided Multiple Instance Learning for Whole Slide Image Classification

Multiple Instance Learning (MIL) is widely used in analyzing histopathol...
research
01/19/2015

Instance Significance Guided Multiple Instance Boosting for Robust Visual Tracking

Multiple Instance Learning (MIL) recently provides an appealing way to a...
research
11/14/2020

MP-Boost: Minipatch Boosting via Adaptive Feature and Observation Sampling

Boosting methods are among the best general-purpose and off-the-shelf ma...
research
12/24/2022

Unsupervised Instance and Subnetwork Selection for Network Data

Unlike tabular data, features in network data are interconnected within ...
research
02/06/2014

Dissimilarity-based Ensembles for Multiple Instance Learning

In multiple instance learning, objects are sets (bags) of feature vector...
research
10/02/2019

Learning Maximally Predictive Prototypes in Multiple Instance Learning

In this work, we propose a simple model that provides permutation invari...
research
05/31/2020

Theory and Algorithms for Shapelet-based Multiple-Instance Learning

We propose a new formulation of Multiple-Instance Learning (MIL), in whi...

Please sign up or login with your details

Forgot password? Click here to reset