
Address Instancelevel Label Prediction in Multiple Instance Learning
Multiple Instance Learning (MIL) is concerned with learning from bags of...
read it

Kernel Mean Embedding of Instancewise Predictions in Multiple Instance Regression
In this paper, we propose an extension to an existing algorithm (instanc...
read it

ConfidenceConstrained Maximum Entropy Framework for Learning from MultiInstance Data
Multiinstance data, in which each object (bag) contains a collection of...
read it

Multiple Instance Learning by Discriminative Training of Markov Networks
We introduce a graphical framework for multiple instance learning (MIL) ...
read it

Twostage Training for Learning from Label Proportions
Learning from label proportions (LLP) aims at learning an instancelevel...
read it

Learning with Sets in Multiple Instance Regression Applied to Remote Sensing
In this paper, we propose a novel approach to tackle the multiple instan...
read it

Bag Reference Vector for Multiinstance Learning
Multiinstance learning (MIL) has a wide range of applications due to it...
read it
Dynamic Programming for Instance Annotation in Multiinstance Multilabel Learning
Labeling data for classification requires significant human effort. To reduce labeling cost, instead of labeling every instance, a group of instances (bag) is labeled by a single bag label. Computer algorithms are then used to infer the label for each instance in a bag, a process referred to as instance annotation. This task is challenging due to the ambiguity regarding the instance labels. We propose a discriminative probabilistic model for the instance annotation problem and introduce an expectation maximization framework for inference, based on the maximum likelihood approach. For many probabilistic approaches, bruteforce computation of the instance label posterior probability given its bag label is exponential in the number of instances in the bag. Our key contribution is a dynamic programming method for computing the posterior that is linear in the number of instances. We evaluate our methods using both benchmark and real world data sets, in the domain of bird song, image annotation, and activity recognition. In many cases, the proposed framework outperforms, sometimes significantly, the current stateoftheart MIML learning methods, both in instance label prediction and bag label prediction.
READ FULL TEXT
Comments
There are no comments yet.