One pass learning for generalized classifier neural network

Özyildirim, Buse Melis; Avcı, Mutlu

doi:10.1016/j.neunet.2015.10.008

Cited by 20 publications

(6 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The discriminant boundary of the PNN is close to the Bayesian optimal decision surface. RBFNN uses RBF to construct the space of hidden layers to make the input data linearly separable [36]. GS-SVM uses a kernel function to map low-dimensional data and constructs a hyperplane to make the data separable.…”

Section: Taste Sensation Evaluationmentioning

confidence: 99%

Taste sensation evaluation for an electronic tongue based on an optimized computational model of taste pathways

Zheng¹,

Shi²,

Xia³

et al. 2022

Meas. Sci. Technol.

View full text Add to dashboard Cite

Effective evaluation of taste sensation can be achieved by analyzing electronic tongue (e-tongue) data. Research on taste sensation evaluation of the e-tongue by nerve conduction mechanisms is limited, which affects the evaluation performance of the e-tongue. Therefore, in this paper, a method for evaluating the taste sensation of the e-tongue based on human taste conduction mechanisms, computational model of taste pathways (CMTP), is proposed. However, the limited physiological basis of the CMTP parameters influences the evaluation results. To achieve excellent evaluation performance, a parameter optimization algorithm using Hebbian and habituation learning rules is used to optimize the CMTP parameters. The effectiveness of the optimized results is demonstrated by the improvement in the dynamic characteristics of the CMTP. Next, the optimized CMTP (OCMTP) is used for pattern recognition and sweetness evaluation of the four taste substances. The results showed, first, that the dynamic characteristics (including 1/f characteristics and synchronization) of the OCMTP are improved, and the bionics of the OCMTP is enhanced. The optimized results are effective. Second, compared with the recognition results of four taste substances by the unoptimized CMTP (UCMTP), signal preprocessing methods, and multiclass classification models, the best classification accuracy of 95.38%, the best Kappa coefficient of 93.83%, and the best F1-score of 96.10% are acquired by the OCMTP. Finally, compared with the sweetness evaluation results of the UCMTP, signal preprocessing methods, and multiple evaluation models, the best evaluation performance, including an RMSE of 0.1643 and R2 of 0.9785, is obtained using the OCMTP. In conclusion, effective evaluation of taste sensation can be achieved by the OCMTP and an e-tongue.

show abstract

Section: Taste Sensation Evaluationmentioning

confidence: 99%

Taste sensation evaluation for an electronic tongue based on an optimized computational model of taste pathways

Zheng¹,

Shi²,

Xia³

et al. 2022

Meas. Sci. Technol.

View full text Add to dashboard Cite

show abstract

“…Tidak seperti jaringan syaraf tiruan berbasis radial basis function lainnya, L-GCNN menggunakan fungsi logarithmic sehingga dapat mengurangi jumlah iterasi untuk mencapai error minimal. Metode ini memiliki 5 layer utama yaitu input, pattern, summation, normalization dan output layer [14].…”

Section: Logarithmic Learning For Generalized Classifier Neural Netwounclassified

Penggunaan Dual Sensor Leap Motion Controller untuk Sistem Isyarat Bahasa Indonesia (SIBI)

Nugraha

Murti²,

Khotimah³

2019

INFORMAL

View full text Add to dashboard Cite

Bahasa isyarat adalah bahasa yang dibentuk dengan kombinasi gerakan jari, tangan, tubuh serta ekspresi wajah yang digunakan oleh para penyandang disabilitas seperti tunarungu dan tunawicara. Salah satu pengenalan bahasa isyarat tersebut adalah pengenalan dengan menggunakan teknologi sensor Leap Motion Controller (LMC). Akan tetapi pada penerapannya, sensor LMC tidak selalu dapat mengenali semua jari dengan baik. Hal ini dikarenakan pembentukan dari bahasa isyarat yang saling bersentuhan antar jari, jari yang dilipat, tersembunyi ataupun terhalang oleh jari lainnya. Sehingga pada penelitian ini diusulkan penggunaan dual sensor LMC yang bertujuan untuk mencegah dan meminimalisir adanya kesalahan dalam mengenal bahasa isyarat dari jari yang dibentuk. Tahapan proses yang dilakukan adalah mendesain tata letak sensor LMC dan mengkombinasikan data fitur dari masing-masing sensor LMC baik yang bersifat statis dan dinamis yang kemudian diklasifikasikan menggunakan metode L-GCNN. Dari hasil pengujian yang dilakukan, penggunaan dual sensor LMC ini dapat mengenal isyarat SIBI yang diberikan. Skema yang terbaik adalah dengan mengacu pada tata letak multi sensor yang saling berhadapan membentuk sudut 120 derajat dengan tingkat akurasi sebesar 86,42%, dengan rata-rata akurasi dari keseluruhan skenario sebesar 84,36%. Adanya desain tata letak sensor LMC dan kombinasi penggabungan data fitur masing-masing sensor dapat meningkatkan akurasi pengenalan dengan rata-rata peningkatan mencapai 5,35%.

show abstract

“…For the convenience of description, the autoencoder model implemented by multi-layer perceptrons will be referred to as AE-MLP and the autoencoder model implemented by residual networks will be referred to as AE-RES in the remainder of this paper. The AE-MLP model needs to be trained in the iterative mode, while the AE-RES model can be trained in the one-pass learning mode [52,53]. (2) The high-dimensional features extracted by the autoencoders are reduced to 2D using the uniform manifold approximate and projection (UMAP) algorithm [54], which is a commonly used dimensionality reduction method in life sciences [55,56].…”

Section: Introductionmentioning

confidence: 99%

An Unsupervised Classification Algorithm for Heterogeneous Cryo-EM Projection Images Based on Autoencoders

Wang

Lin

et al. 2023

IJMS

View full text Add to dashboard Cite

Heterogeneous three-dimensional (3D) reconstruction in single-particle cryo-electron microscopy (cryo-EM) is an important but very challenging technique for recovering the conformational heterogeneity of flexible biological macromolecules such as proteins in different functional states. Heterogeneous projection image classification is a feasible solution to solve the structural heterogeneity problem in single-particle cryo-EM. The majority of heterogeneous projection image classification methods are developed using supervised learning technology or require a large amount of a priori knowledge, such as the orientations or common lines of the projection images, which leads to certain limitations in their practical applications. In this paper, an unsupervised heterogeneous cryo-EM projection image classification algorithm based on autoencoders is proposed, which only needs to know the number of heterogeneous 3D structures in the dataset and does not require any labeling information of the projection images or other a priori knowledge. A simple autoencoder with multi-layer perceptrons trained in iterative mode and a complex autoencoder with residual networks trained in one-pass learning mode are implemented to convert heterogeneous projection images into latent variables. The extracted high-dimensional features are reduced to two dimensions using the uniform manifold approximation and projection dimensionality reduction algorithm, and then clustered using the spectral clustering algorithm. The proposed algorithm is applied to two heterogeneous cryo-EM datasets for heterogeneous 3D reconstruction. Experimental results show that the proposed algorithm can effectively extract category features of heterogeneous projection images and achieve high classification and reconstruction accuracy, indicating that the proposed algorithm is effective for heterogeneous 3D reconstruction in single-particle cryo-EM.

show abstract

One pass learning for generalized classifier neural network

Cited by 20 publications

References 16 publications

Taste sensation evaluation for an electronic tongue based on an optimized computational model of taste pathways

Taste sensation evaluation for an electronic tongue based on an optimized computational model of taste pathways

Penggunaan Dual Sensor Leap Motion Controller untuk Sistem Isyarat Bahasa Indonesia (SIBI)

An Unsupervised Classification Algorithm for Heterogeneous Cryo-EM Projection Images Based on Autoencoders

Contact Info

Product

Resources

About