Supporting Deep Neural Network Safety Analysis and Retraining Through Heatmap-Based Unsupervised Learning

2021 ◽  
pp. 1-17
Author(s):  
Hazem Fahmy ◽  
Fabrizio Pastore ◽  
Mojtaba Bagherzadeh ◽  
Lionel Briand
2018 ◽  
Vol 27 (6) ◽  
pp. 3049-3063 ◽  
Author(s):  
Zhizhong Han ◽  
Zhenbao Liu ◽  
Chi-Man Vong ◽  
Yu-Shen Liu ◽  
Shuhui Bu ◽  
...  

2020 ◽  
Vol 7 (4) ◽  
pp. 727
Author(s):  
Larasati Larasati ◽  
Wisnu Ananta Kusuma ◽  
Annisa Annisa

<p class="Abstrak"><em>Drug repositioning</em> adalah penggunaan senyawa obat yang sudah lolos uji sebelumnya untuk mengatasi penyakit baru selain penyakit awal obat tersebut ditujukan. <em>Drug repositioning </em>dapat dilakukan dengan memprediksi interaksi senyawa obat dengan protein penyakit yang bereaksi positif. Salah satu tantangan dalam prediksi interaksi senyawa dan protein adalah masalah ketidakseimbangan data. <em>Deep semi-supervised learning </em>dapat menjadi alternatif untuk menangani model prediksi dengan data yang tidak seimbang. Proses <em>pre-training </em>berbasis <em>unsupervised learning</em> pada <em>deep semi-supervised learning </em>dapat merepresentasikan input dari <em>unlabeled data</em> (data mayoritas) dengan baik dan mengoptimasi inisialisasi bobot pada <em>classifier</em>. Penelitian ini mengimplementasikan <em>Deep Belief Network</em> (DBN) sebagai <em>pre-training</em> dan <em>Deep Neural Network</em> (DNN) sebagai <em>classifier</em>. Data yang digunakan pada penelitian ini adalah <em>dataset</em> ion channel, GPCR, dan nuclear receptor yang bersumber dari pangkalan data KEGG BRITE, BRENDA, SuperTarget, dan DrugBank. Hasil penelitian ini menunjukkan pada <em>dataset</em> tersebut, <em>pre-training</em> berupa ekstraksi fitur memberikan efek optimasi dilihat dari peningkatan performa model DNN pada akurasi (3-4.5%), AUC (4.5%), <em>precision</em><em> </em>(5.9-6%), dan F-measure (3.8%).</p><p class="Abstrak"> </p><p class="Abstrak"><em><strong>Abstract</strong></em></p><p class="Abstract"><em>Drug repositioning is the reuse of an existing drug to treat a new disease other than its original medical indication. Drug repositioning can be done by predicting the interaction of drug compounds with disease proteins that react positively. One of the challenges in predicting the interaction of compounds and proteins is imbalanced data. Deep semi-supervised learning can be an alternative to handle prediction models with imbalanced data. The unsupervised learning based pre-training process in deep semi-supervised learning can represent input from unlabeled data (majority data) properly and optimize initialization of weights on the classifier. This study implements the Deep Belief Network (DBN) as a pre-training with Deep Neural Network (DNN) as a classifier. The data used in this study are ion channel, GPCR, and nuclear receptor dataset sourced from KEGG BRITE, BRENDA, SuperTarget, and DrugBank databases. The results of this study indicate that pre-training as feature extraction had an optimization effect. This can be seen from DNN performance improvement in accuracy (3-4.5%), AUC (4.5%), precision (5.9-6%), and F-measure (3.8%).<strong></strong></em></p><p class="Abstrak"><em><strong><br /></strong></em></p>


2021 ◽  
Author(s):  
Bardia Esmaeili ◽  
Alireza Akhavanpour ◽  
Mohammad Sabokrou

2020 ◽  
Vol 105 (sp1) ◽  
Author(s):  
Zixin Liu ◽  
Mingxing Ling ◽  
Ting Zhu ◽  
Deru Xu

2021 ◽  
Author(s):  
Anna Metzger ◽  
Matteo Toscani

AbstractWhen touching the surface of an object, its spatial structure translates into a vibration on the skin. The perceptual system evolved to translate this pattern into a representation that allows to distinguish between different materials. Here we show that perceptual haptic representation of materials emerges from efficient encoding of vibratory patterns elicited by the interaction with materials. We trained a deep neural network with unsupervised learning (Autoencoder) to reconstruct vibratory patterns elicited by human haptic exploration of different materials. The learned compressed representation (i.e. latent space) allows for classification of material categories (i.e. plastic, stone, wood, fabric, leather/wool, paper, and metal). More importantly, distances between these categories in the latent space resemble perceptual distances, suggesting a similar coding. We could further show, that the temporal tuning of the emergent latent dimensions is similar to properties of human tactile receptors.


Author(s):  
David T. Wang ◽  
Brady Williamson ◽  
Thomas Eluvathingal ◽  
Bruce Mahoney ◽  
Jennifer Scheler

Sign in / Sign up

Export Citation Format

Share Document