Pool-Based Active Learning with Query Construction

Zhang, Shanhong; Yin, Jian; Guo, Weizhao

doi:10.1007/978-3-642-25664-6_2

Cited by 4 publications

(3 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In the case of small datasets, it is difficult to achieve satisfactory performances. However, active learning techniques, or also called query learning techniques, stand out to solve the issue of data scarcity [8]. These methods are pool-based approaches, where the dataset is organized in terms of pools and, iteratively, samples of the pools are chosen following a query strategy.…”

Section: ) Active Learning Approachmentioning

confidence: 99%

Tree Detection and Localization Approach for UAV-based Forest Inspection

Battseren

Mohamed

Diego

et al. 2022

ESS

View full text Add to dashboard Cite

This paper addressed an UAV-based solution for forest inspection from the aspect of data processing. Insect infected trees can be identified from aerial imagery. The forest inspection data processing divided into two main phases, which are tree detection and tree localization. For the tree detection task, the Deep Learning-based approach has been used. In this study, a conventional and active learning method have been used for the training. In order to map the detected trees from the video frame to geographical locations, a custom geo-localization algorithm has been developed. The proposed system loads the inspection video as an input and calculates and outputs the detected trees’ GPS coordinates. In the calculation Digital Elevation Model and camera parameters are used in addition to drone geo-data to calculate the tree positions.

show abstract

Section: ) Active Learning Approachmentioning

confidence: 99%

Tree Detection and Localization Approach for UAV-based Forest Inspection

Battseren

Mohamed

Diego

et al. 2022

ESS

View full text Add to dashboard Cite

show abstract

“…Active learning (AL) aims to select the most valuable samples from a pool of unlabeled samples using a query strategy. Among them, the pool-based active learning (Pool-Based AL) [19][20][21] sample selection method has been widely studied. Most previous active learning methods use a single query strategy to select samples and train models in the same domain.…”

Section: Introductionmentioning

confidence: 99%

Active domain adaptation method for label expansion problem

Zhang

Qinle

et al. 2022

Proceedings of the Institution of Mechanical Engineers, Part O:

View full text Add to dashboard Cite

Over the past few years, cross-domain fault detection methods based on unsupervised domain adaptation (UDA) have gradually matured. However, existing methods usually assume that the source and target domains have the same label domain space, but ignore the problem of label expansion in the target domain. The source domain of such problems lacks transferable knowledge of newly added health categories, so the domain invariant features extracted by the UDA model only have a large correlation with the source domain health categories, but lack the key features to distinguish the newly added health categories. We found that most of the diagnostic results of this type of samples are distributed at the decision boundary of the source domain health category, and this special distribution means that the newly added health category samples have a high amount of information. Therefore, this paper considers using active learning to select samples of newly added health categories in the target domain to assist model training, and proposes an active domain adaptation intelligent fault detection framework LDE-ADA to deal with the label expansion problem. Finally, on the rotating machinery dataset, the analysis and comparison are carried out through six transfer tasks. The results show that when there is one new health category, the accuracy of LDE-ADA will increase by about 9.39% in the case of labeling three samples per round and training for 20 rounds. Experiments show that this method is an effective method to deal with the label expansion problem.

show abstract

“…Aprendizagem ativa é um importante método para resolver o problema de poucos dados anotados em abordagens de aprendizagem de máquina [120]. Quando há falta de dados na etapa de aprendizagem de um modelo, amostras são anotadas por especialistas humanos, o que é um progresso caro e demorado.…”

Section: Aprendizagem Ativaunclassified

Habilitando Anotações De Dados Autônomos: Uma Abordagem De Aprendizado Por Reforço Com Humano No Loop

CRUZ¹

View full text Add to dashboard Cite

Pool-Based Active Learning with Query Construction

Cited by 4 publications

References 9 publications

Tree Detection and Localization Approach for UAV-based Forest Inspection

Tree Detection and Localization Approach for UAV-based Forest Inspection

Active domain adaptation method for label expansion problem

Habilitando Anotações De Dados Autônomos: Uma Abordagem De Aprendizado Por Reforço Com Humano No Loop

Contact Info

Product

Resources

About