Graph-based PU learning for binary and multiclass classification without class prior

Yoo, Junehee; Kim, Jung-Hun; Yoon, Hoyoung; Kim, Geonsoo; Jang, Changwon; Kang, U

doi:10.1007/s10115-022-01702-8

Cited by 4 publications

(5 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The results are presented in Tables VI and VII, respectively. It can be seen that the generating ability of our proposed style loss is significantly better than that of 3C-GAN [32] and KEGNET [31], with respect to solving the problem of source domain generation in UniDA without source data. In addition, our style loss maintains a more prominent and uniform performance in perclass accuracy.…”

Section: ) Comparison Of Different Data Diversity Generation Schemesmentioning

confidence: 94%

“…Thus, D f = {arg max x p(x | y, z)}. Based on the Bayesian theory [31], [53], arg max x p(x | y, z) can be expressed as follows:…”

Section: B Source Data Generationmentioning

confidence: 99%

“…Specifically, a discriminator is introduced to match the distributions between the target samples and the generated source samples through the use of adversarial training. The second, a decoder loss in KEGNET [31], produces similar data points for each class and increases the pairwise distance between sampled data points. We compare our proposed style loss in data diversity module with these two methods on RSSCN7 → UCM and AID→NWPU.…”

Section: ) Comparison Of Different Data Diversity Generation Schemesmentioning

confidence: 99%

“…Therefore, generating synthetic source domain data from the pretrained model is the first problem to be solved. There are recent works for distilling a network's knowledge by a small dataset [30] or no observable data [31]. It is worth noting that we cannot use GANs to directly generate artificial data (similar to [32]) because the core of UniDA without source data is to restore the category distribution (including the shared label set and the private label set) from the pretrained model.…”

mentioning

confidence: 99%

See 3 more Smart Citations

Universal Domain Adaptation for Remote Sensing Image Scene Classification

Shi

Yuan

et al. 2023

IEEE Trans. Geosci. Remote Sensing

View full text Add to dashboard Cite

The domain adaptation (DA) approaches available to date are usually not well suited for practical DA scenarios of remote sensing image classification since these methods (such as unsupervised DA) rely on rich prior knowledge about the relationship between label sets of source and target domains, and source data are often not accessible due to privacy or confidentiality issues. To this end, we propose a practical universal DA (UniDA) setting for remote sensing image scene classification that requires no prior knowledge on the label sets. Furthermore, a novel UniDA method without source data is proposed for cases when the source data are unavailable. The architecture of the model is divided into two parts: the source data generation stage and the model adaptation stage. The first stage estimates the conditional distribution of source data from the pretrained model using the knowledge of class separability in the source domain and then synthesizes the source data. With this synthetic source data in hand, it becomes a UniDA task to classify a target sample correctly if it belongs to any category in the source label set or mark it as "unknown" otherwise. In the second stage, a novel transferable weight that distinguishes the shared and private label sets in each domain promotes the adaptation in the automatically discovered shared label set and recognizes the "unknown" samples successfully. Empirical results show that the proposed model is effective and practical for remote sensing image scene classification, regardless of whether the source data are available or not. The code is available at

show abstract

Section: ) Comparison Of Different Data Diversity Generation Schemesmentioning

confidence: 94%

“…Thus, D f = {arg max x p(x | y, z)}. Based on the Bayesian theory [31], [53], arg max x p(x | y, z) can be expressed as follows:…”

Section: B Source Data Generationmentioning

confidence: 99%

Section: ) Comparison Of Different Data Diversity Generation Schemesmentioning

confidence: 99%

mentioning

confidence: 99%

See 2 more Smart Citations

Universal Domain Adaptation for Remote Sensing Image Scene Classification

Shi

Yuan

et al. 2023

IEEE Trans. Geosci. Remote Sensing

View full text Add to dashboard Cite

show abstract

“…On the other hand, PU-LP, which has a time complexity of O(n 2 logn), [Jaemin et al 2022] is a graph-based algorithm that utilizes a similarity matrix created using the k-Nearest Neighbors approach. It identifies the unlabeled nodes with the lowest similarity scores and designates them as reliable negative examples.…”

Section: Positive Unlabeled Learningmentioning

confidence: 99%

Positive Unlabeled Learning: Adapting NMF for text classification

Nunes,

Faleiros,

Rossi

2023

Anais Do XX Encontro Nacional De Inteligência Artificial E Computacional (ENIAC 2023)

View full text Add to dashboard Cite

Due to the overwhelming data generation that surpasses human evaluation capacity, manually labeling data for training machine learning models is becoming increasingly impractical. This article focuses on analyzing techniques to address the challenges of Positive Unlabeled Learning (PUL). To this end, we propose structural adaptations to the Non-Negative Matrix Factorization (NMF) algorithm, specifically tailored for PU data (NMFPUL). We compare NMFPUL with state-of-the-art techniques to identify improvements in the performance of textual data classification. Our study reveals that NMFPUL consistently outperforms most baseline algorithms across diverse document collections even with a limited number of labeled documents, and mainly on these situations.

show abstract