Zero initialized active learning with spectral clustering using Hungarian method

Papp Dávid: Zero initialized active learning with spectral clustering using Hungarian method. In: Acta cybernetica, (25) 2. pp. 401-419. (2021)

[thumbnail of cybernetica_025_numb_002_401-419.pdf] Cikk, tanulmány, mű

Download (628kB)


Supervised machine learning tasks often require a large number of labeled training data to set up a model, and then prediction - for example the classification - is carried out based on this model. Nowadays tremendous amount of data is available on the web or in data warehouses, although only a portion of those data is annotated and the labeling process can be tedious, expensive and time consuming. Active learning tries to overcome this problem by reducing the labeling cost through allowing the learning system to iteratively select the data from which it learns. In special case of active learning, the process starts from zero initialized scenario, where the labeled training dataset is empty, and therefore only unsupervised methods can be performed. In this paper a novel query strategy framework is presented for this problem, called Clustering Based Balanced Sampling Framework (CBBSF), which is not only select the initial labeled training dataset, but uniformly selects the items among the categories to get a balanced labeled training dataset. The framework includes an assignment technique to implicitly determine the class membership probabilities. Assignment solution is updated during CBBSF iterations, hence it simulates supervised machine learning more accurately as the process progresses. The proposed Spectral Clustering Based Sampling (SCBS) query startegy realizes the CBBSF framework, and therefore it is applicable in the special zero initialized situation. This selection approach uses ClusterGAN (Clustering using Generative Adversarial Networks) integrated in the spectral clustering algorithm and then it selects an unlabeled instance depending on the class membership probabilities. Global and local versions of SCBS were developed, furthermore, most confident and minimal entropy measures were calculated, thus four different SCBS variants were examined in total. Experimental evaluation was conducted on the MNIST dataset, and the results showed that SCBS outperforms the state-of-the-art zero initialized active learning query strategies.

Item Type: Article
Journal or Publication Title: Acta cybernetica
Date: 2021
Volume: 25
Number: 2
ISSN: 0324-721X
Page Range: pp. 401-419
Language: English
Publisher: University of Szeged, Institute of Informatics
Place of Publication: Szeged
Event Title: Conference of PhD Students in Computer Science (12.) (2020) (Szeged)
Related URLs:
DOI: 10.14232/actacyb.288006
Uncontrolled Keywords: Algoritmus, Programozás
Additional Information: Bibliogr.: p. 416-419. ; ill. ; összefoglalás angol nyelven
Subjects: 01. Natural sciences
01. Natural sciences > 01.02. Computer and information sciences
Date Deposited: 2022. May. 12. 15:34
Last Modified: 2022. May. 12. 15:34

Actions (login required)

View Item View Item