Efficient rejection strategies for prototype-based classification

Fischer, Lydia; Hammer, Barbara; Wersing, Heiko

doi:10.1016/j.neucom.2014.10.092

Cited by 32 publications

(28 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It has been analysed lately that this value serves as an efficient estimation of a confidence for classification with rejection showing similar performance as an explicit probabilistic modelling but at lower computational costs [43]. The calculation of the RelSim for unlabelled data points x is based on the estimated class label with respect to the model, i. e. d…”

Section: Learning Vector Quantizationmentioning

confidence: 99%

Combining offline and online classifiers for life-long learning

Fischer

Hammer

Wersing³

2015

2015 International Joint Conference on Neural Networks (IJCNN)

View full text Add to dashboard Cite

Abstract-One of the greatest challenges of life-long learning architectures is how to efficiently and reliably cope with the stability-plasticity dilemma. We propose an extension of a flexible system combining a static offline classifier and an incremental online classifier that is well suited for life-long learning scenarios. The pre-trained offline classifier preserves ground knowledge that should be respected during training, while the online classifier enables learning of new or specific information encountered during use. The combination is realised by a dynamic classifier selection strategy based on confidences of both ingredients. We report exemplary results of this architecture for the case of learning vector quantization (LVQ) for several data sets, thereby including an extensive comparison to alternative state of the art algorithms for incremental learning such as incremental generalised LVQ and the support vector machine.

show abstract

Section: Learning Vector Quantizationmentioning

confidence: 99%

Combining offline and online classifiers for life-long learning

Fischer

Hammer

Wersing³

2015

2015 International Joint Conference on Neural Networks (IJCNN)

View full text Add to dashboard Cite

show abstract

“…The latter case occurs if we observe overlapping class distributions in the training data, i.e. ambiguous regions near the classs borders whereas an example for aleatoric uncertainty are outliers in data [96,97], see Fig. 4.…”

Section: Reject or Classify -Secure Classificationmentioning

confidence: 99%

Can Learning Vector Quantization be an Alternative to SVM and Deep Learning? - Recent Trends and Advanced Variants of Learning Vector Quantization for Classification Learning

Villmann

Bohnsack

Kaden

2016

Journal of Artificial Intelligence and Soft Computing Research

View full text Add to dashboard Cite

Learning vector quantization (LVQ) is one of the most powerful approaches for prototype based classification of vector data, intuitively introduced by Kohonen. The prototype adaptation scheme relies on its attraction and repulsion during the learning providing an easy geometric interpretability of the learning as well as of the classification decision scheme. Although deep learning architectures and support vector classifiers frequently achieve comparable or even better results, LVQ models are smart alternatives with low complexity and computational costs making them attractive for many industrial applications like intelligent sensor systems or advanced driver assistance systems.Nowadays, the mathematical theory developed for LVQ delivers sufficient justification of the algorithm making it an appealing alternative to other approaches like support vector machines and deep learning techniques.This review article reports current developments and extensions of LVQ starting from the generalized LVQ (GLVQ), which is known as the most powerful cost function based realization of the original LVQ. The cost function minimized in GLVQ is an soft-approximation of the standard classification error allowing gradient descent learning techniques. The GLVQ variants considered in this contribution, cover many aspects like bordersensitive learning, application of non-Euclidean metrics like kernel distances or divergences, relevance learning as well as optimization of advanced statistical classification quality measures beyond the accuracy including sensitivity and specificity or area under the ROC-curve.According to these topics, the paper highlights the basic motivation for these variants and extensions together with the mathematical prerequisites and treatments for integration into the standard GLVQ scheme and compares them to other machine learning approaches. For detailed description and mathematical theory behind all, the reader is referred to the respective original articles.Thus, the intention of the paper is to provide a comprehensive overview of the stateof-the-art serving as a starting point to search for an appropriate LVQ variant in case of a given specific classification problem as well as a reference to recently developed variants and improvements of the basic GLVQ scheme.

show abstract

“…NPC algorithms decrease classification time by reducing the number of instances in the data set [22]. Since the accuracy of such classification techniques is very important, Deng, et al [25] apply transfer learning to prototype-based fuzzy clustering (PFC) in order to solve the problem of limitation and scarcity of data for clustering task, and Fischer, et al [36] present simple and efficient reject options for prototype-based classification to reach a reliable decision.…”

Section: Introductionmentioning

confidence: 99%

“…, expresses preferences of a representativeuser k on the item j and , -closeness of the user to the representative user k. We normalize , (̂, ) to get , ′ (̂, ′) (36) We give the according to the rules:…”

mentioning

confidence: 99%

Fuzzy prototype classifier based on items and its application in recommender system

Mei¹,

Gong²,

Li³

2017

IJCIS

View full text Add to dashboard Cite

Currently, recommender systems (RS) are incorporating implicit information from social circle of the Internet. The implicit social information in human mind is not easy to reflect in appropriate decision making techniques. This paper consists of 2 contributions. First, we develop an item-based prototype classifier (IPC) in which a prototype represents a social circle's preferences as a pattern classification technique. We assume the social circle which distinguishes with others by the items their members like. The prototype structure of the classifier is defined by two2-dimensional matrices. We use information gain and OWA aggregator to construct a feature space. The item-based classifier assigns a new item to some prototypes with different prototypicalities. We reform a typical data set-Iris data set in UCI Machine Learning Repository to verify our fuzzy prototype classifier. The second proposition of this paper is to give the application of IPC in recommender system to solve new item cold-start problems. We modify the dataset of MovieLens to perform experimental demonstrations of the proposed ideas.

show abstract

Efficient rejection strategies for prototype-based classification

Cited by 32 publications

References 30 publications

Combining offline and online classifiers for life-long learning

Combining offline and online classifiers for life-long learning

Can Learning Vector Quantization be an Alternative to SVM and Deep Learning? - Recent Trends and Advanced Variants of Learning Vector Quantization for Classification Learning

Fuzzy prototype classifier based on items and its application in recommender system

Contact Info

Product

Resources

About