Instance Selection Optimization for Neural Network Training

Kordos, Mirosław

doi:10.1007/978-3-319-39378-0_52

Search citation statements

Order By: Relevance

Paper Sections

Select...

Instance Selection Before Network Learning1

Citation Types

Supporting

Mentioning

Contrasting

Year Published

2017

2018

Publication Types

Select...

Other2

Article1

Relationship

Self Cite1

Independent2

Authors

Journals

Cited by 3 publications

(1 citation statement)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the standard CNN an instance will get removed if it has the same class as more then k/2 of its neighbors, that is m = 5 in that case. If we increase m say to 8 then the instance will get removed if it has the same class as more then m = 8 of its neighbors, so the selection will be less aggressive and some of the instance situated close to class boundaries that with m = 5 would get removed will be kept [17]. Again the compression will be weaker but the accuracy will be higher.…”

Section: Instance Selection Before Network Learningmentioning

confidence: 99%

Data Selection for Neural Networks

Kordos

2017

Self Cite

View full text Add to dashboard Cite

Abstract. Several approaches to joined feature and instance selection in neural network leaning are discussed and experimentally evaluated in respect to classification accuracy and dataset compression, considering also their computational complexity. These include various versions of feature and instance selection prior to the network learning, the selection embedded in the neural network and hybrid approaches, including solutions developed by us. The advantages and disadvantages of each approach are discussed and some possible improvements are proposed.

show abstract

Section: Instance Selection Before Network Learningmentioning

confidence: 99%