A self-adaptive local metric learning method for classification

Taheri, Mahsa; Moslehi, Zahra; Mirzaei, Abdolreza; Safayani, Mehran

doi:10.1016/j.patcog.2019.106994

Cited by 13 publications

(3 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The framework can be naturally derived from the global, multi-task and local measure learning problems. Recently, a new supervised self-adaptive local ML method (SA-LM 2 ) is proposed in [18]. It adaptively adjust local neighborhood to construct the proper distance to separate the distinct categories.…”

Section: A Metric Learningmentioning

confidence: 99%

“…With the use of the sample pairwise distance defined in Eq. ( 14), this paper co-opts the advantage of the ML method in [18] and further embeds a self-adaptive loss function into 8); these initialized K +1 AEs to produce the features with decent discernibility.…”

Section: B Training Sadaementioning

confidence: 99%

“…In order to simultaneously value the local information and extract the discriminative features that will constitute decent representations of the medical images, this paper coopts the advantages of a self-adaptive local ML strategy [18] and proposes a self-adaptive discriminative autoencoder (SADAE). The framework of this approach consists of K local autoencoders (AEs) to learn K subspaces implying the diverse local distribution of data and a global autoencoder (AE) to restrict the spatial scale of the learned features.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A Self-Adaptive Discriminative Autoencoder for Medical Applications

Shang

et al. 2022

IEEE Trans. Circuits Syst. Video Technol.

View full text Add to dashboard Cite

Computer aided diagnosis (CAD) systems play an essential role in the early detection and diagnosis of developing disease for medical applications. In order to obtain the highly recognizable representation for the medical images, a self-adaptive discriminative autoencoder (SADAE) is proposed in this paper. The proposed SADAE system is implemented under a deep metric learning framework which consists of K local autoencoders, employed to learn the K subspaces that represent the diverse distribution of the underlying data, and a global autoencoder to restrict the spatial scale of the learned representation of images. Such community of autoencoders is aided by a self-adaptive metric learning method that extracts the discriminative features to recognize the different categories in the given images. The quality of the extracted features by SADAE is compared against that of those extracted by other state-of-the-art deep learning and metric learning methods on five popular medical image data sets. The experimental results demonstrate that the medical image recognition results gained by SADAE are much improved over those by the alternatives.

show abstract

Section: A Metric Learningmentioning

confidence: 99%

Section: B Training Sadaementioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation