Learning from Few Samples with Memory Network

Zhang, Shufei; Huang, Kaizhu; Zhang, Rui; Hussain, Amir

doi:10.1007/s12559-017-9507-z

Cited by 16 publications

(4 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For this reason, we plan to explore the combination of advanced feature extraction technique [52] and MTL strategy. Furthermore, obtaining a joint recognition with a small amount of label data is also valuable, which might be achieved by integrating the MTL with the approach of learning from few examples [53].…”

Section: Discussionmentioning

confidence: 99%

Context-Aware Human Activity and Smartphone Position-Mining with Motion Sensors

et al. 2019

Self Cite

View full text Add to dashboard Cite

Today's smartphones are equipped with embedded sensors, such as accelerometers and gyroscopes, which have enabled a variety of measurements and recognition tasks. In this paper, we jointly investigate two types of recognition problems in a joint manner, e.g., human activity recognition and smartphone on-body position recognition, in order to enable more robust context-aware applications. So far, these two problems have been studied separately without considering the interactions between each other. In this study, by first applying a novel data preprocessing technique, we propose a joint recognition framework based on the multi-task learning strategy, which can reduce computational demand, better exploit complementary information between the two recognition tasks, and lead to higher recognition performance. We also extend the joint recognition framework so that additional information, such as user identification with biometric motion analysis, can be offered. We evaluate our work systematically and comprehensively on two datasets with real-world settings. Our joint recognition model achieves the promising performance of 0.9174 in terms of F1-score for user identification on the benchmark RealWorld Human Activity Recognition (HAR) dataset. On the other hand, in comparison with the conventional approach, the proposed joint model is shown to be able to improve human activity recognition and position recognition by 5.1% and 9.6% respectively.

show abstract

Section: Discussionmentioning

confidence: 99%

Context-Aware Human Activity and Smartphone Position-Mining with Motion Sensors

et al. 2019

Self Cite

View full text Add to dashboard Cite

show abstract

“…Overall, both models have their strengths and weaknesses, and the choice depends on the specific research question being addressed. [26,[41][42][43][44].…”

Section: ) Spiking Neural Network (Snn) Models Of Hippocampus  Neura...mentioning

confidence: 99%

Uncovering the Neurobiological Consequences of High-Voltage Electrical Field Exposure on the Visual Working Memory of Macaques and Also Using Spiking Neural Network Model

Aliyari,

Hosseinian,

Menhaj

et al. 2023

BCN

View full text Add to dashboard Cite

High-voltage (HV) power transmission lines running near cities and villages can cause severe damage (Mental and physical). Due to the magnetic and electric fields they produce. This study aimed to investigate the effects of high-voltage (HV) electric fields on the spiking neural network model of the brain and biological and behavioral models of visual working memory. To achieve this goal, macaques were studied for their cognitive functions, expression of the NMDA receptor gene, MRI-assisted analysis of brain anatomy, and variations in blood sodium and potassium concentrations. The experimental group of macaques was exposed to a 3kV/m high-voltage field for four hours a day for one month. Computational models were then evaluated using experimental parameters. According to the results, it was observed that being exposed to high-voltage electric fields led to a reduction in the expression of the NMDA receptor gene, as well as a decrease in the levels of Sodium and potassium ions in the blood plasma. Additionally, analysis assisted by MRI showed a decrease in the volume of the hippocampus and amygdala after exposure to the electric field. In conclusion, the results of cognitive, genetic, blood, and MRI tests, along with the spiking neural network model, elucidate the mechanism of the visual working memory deterioration in macaques due to high-voltage electric field exposure.

show abstract

“…Next, to obtain the "generalized Rayleigh quotient" of semi-supervised LDA, the new betweenclass scatter matrix (8) where (9) where i denotes the element subscript of vector j h .…”

Section: Semi-supervised Ldamentioning

confidence: 99%

“…Nevertheless, the sample labeling for RS images is time-consuming, and the robustness of the classifier is worse when the labelled samples are insufficient [7]. But unlabeled samples can be generated in a relatively easier way and their rich characteristic information is helpful for improving the classification performance [8]. Semi-supervised learning can automatically exploits unlabeled samples to improve the learning performance based on labeled samples.…”

Section: Introductionmentioning

confidence: 99%

A Novel Semi-supervised Classification Method Based on Class Certainty of Samples

Gao

Yue

Wang

et al. 2018

Advances in Brain Inspired Cognitive Systems

Self Cite

View full text Add to dashboard Cite

The traditional classification method based on supervised learning classifies remote sensing (RS) images by using sufficient labeled samples. However, the number of labeled samples is limited due to the expensive and time-consuming collection. To effectively utilize the information of unlabeled samples in the learning process, this paper proposes a novel semi-supervised classification method based on class certainty of samples (CCS). First, the class certainty of unlabeled samples obtained based on multi-class SVM is smoothed for robustness. Then, a new semisupervised linear discriminant analysis (LDA) is presented based on class certainty, which improves the separability of samples in the projection subspace. Ultimately, we extend the semi-supervised LDA to nonlinear dimensional reduction by combining class certainty and kernel methods. Furthermore, to assess the effectiveness of proposed method, the nearest neighbor classifier is adopted to classify actual SAR images. The results demonstrate that the proposed method can effectively exploit the information of unlabeled samples and greatly improve the classification effect compared with other state-of-the-art approaches.

show abstract

Learning from Few Samples with Memory Network

Cited by 16 publications

References 20 publications

Context-Aware Human Activity and Smartphone Position-Mining with Motion Sensors

Context-Aware Human Activity and Smartphone Position-Mining with Motion Sensors

Uncovering the Neurobiological Consequences of High-Voltage Electrical Field Exposure on the Visual Working Memory of Macaques and Also Using Spiking Neural Network Model

A Novel Semi-supervised Classification Method Based on Class Certainty of Samples

Contact Info

Product

Resources

About