2015
DOI: 10.1109/tgrs.2015.2406334
|View full text |Cite
|
Sign up to set email alerts
|

Functions of Multiple Instances for Learning Target Signatures

Abstract: The functions of multiple instances (FUMI) approach for learning target and nontarget signatures is introduced. FUMI is a generalization of the multiple-instance learning (MIL) approach for supervised learning. FUMI differs significantly from standard MIL and supervised learning approaches because only data points which are functions of class concepts/signatures are available. In particular, this paper addresses the problem in which data points are convex combinations of target and nontarget signatures. Two al… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
1
1

Citation Types

0
47
0

Year Published

2015
2015
2023
2023

Publication Types

Select...
4
3
1

Relationship

4
4

Authors

Journals

citations
Cited by 48 publications
(47 citation statements)
references
References 51 publications
0
47
0
Order By: Relevance
“…For this reason, WEMI detection tasks adhere well to multiple instance learning (MIL) models. 4,[18][19][20][21][22][23] Multiple instance learning is a variation on supervised learning for problems with uncertain or imprecisely-labeled training data. 18 Instead of pairing each training sample with a class label, MIL methods learn from a set of labeled concepts called "bags".…”
Section: Multiple Instance Ace (Mi-ace)/ Smf (Mi-smf)mentioning
confidence: 99%
“…For this reason, WEMI detection tasks adhere well to multiple instance learning (MIL) models. 4,[18][19][20][21][22][23] Multiple instance learning is a variation on supervised learning for problems with uncertain or imprecisely-labeled training data. 18 Instead of pairing each training sample with a class label, MIL methods learn from a set of labeled concepts called "bags".…”
Section: Multiple Instance Ace (Mi-ace)/ Smf (Mi-smf)mentioning
confidence: 99%
“…4 When the current point is labeled as a target point (i.e., representing a target of interest) the linear combination can be written as…”
Section: Extended Functions Of Multiple Instancesmentioning
confidence: 99%
“…Several previous methods for target characterization have been developed in the literature. The FUnctions of Multiple Instance (FUMI) algorithms [17], [18] learn representative concept from reconstruction error of the uncertainly labeled data. Compared with FUMI, MI-HE learns a discriminative target concept that maximizes the detection response of possible target regions so the estimated signatures are more discriminative for target detection.…”
Section: Introductionmentioning
confidence: 99%