Gaining Visitor Consent for Research III: A Trilingual Posted-Sign Method

2015 ◽  
Vol 58 (4) ◽  
pp. 369-381 ◽  
Author(s):  
Lisa Sindorf ◽  
Joshua Gutwill ◽  
Veronica Garcia-Luis
Keyword(s):  
Sign Method

scholarly journals Deep ConvNet: Non-Random Weight Initialization for Repeatable Determinism, Examined with FSGM

Sensors ◽  
2021 ◽  
Vol 21 (14) ◽  
pp. 4772
Author(s):  
Richard N. M. Rudd-Orthner ◽  
Lyudmila Mihaylova
Keyword(s):  
Cross Validation ◽  
Color Image ◽  
Validation Dataset ◽  
Number Sequence ◽  
Random Weight ◽  
Fast Gradient ◽  
Weight Initialization ◽  
Fitting In ◽  
Sequence Substitution ◽  
Sign Method

A repeatable and deterministic non-random weight initialization method in convolutional layers of neural networks examined with the Fast Gradient Sign Method (FSGM). Using the FSGM approach as a technique to measure the initialization effect with controlled distortions in transferred learning, varying the dataset numerical similarity. The focus is on convolutional layers with induced earlier learning through the use of striped forms for image classification. Which provided a higher performing accuracy in the first epoch, with improvements of between 3–5% in a well known benchmark model, and also ~10% in a color image dataset (MTARSI2), using a dissimilar model architecture. The proposed method is robust to limit optimization approaches like Glorot/Xavier and He initialization. Arguably the approach is within a new category of weight initialization methods, as a number sequence substitution of random numbers, without a tether to the dataset. When examined under the FGSM approach with transferred learning, the proposed method when used with higher distortions (numerically dissimilar datasets), is less compromised against the original cross-validation dataset, at ~31% accuracy instead of ~9%. This is an indication of higher retention of the original fitting in transferred learning.

scholarly journals Digital Signature for data and documents using operating PKI certificates

2021 ◽  
Vol 119 ◽  
pp. 07004
Author(s):  
Adil El Mane ◽  
Younes Chihab ◽  
Redouan Korchiyne
Keyword(s):  
Digital Signature ◽  
Hash Function ◽  
Encryption Algorithm ◽  
Central Theme ◽  
Digital Certificates ◽  
Sign Method

This report represents my researching work for a six-month internship in Lex Persona Enterprise. "The implementation of applications that sign data and files electronically using operating digital certificates" is the central theme of this research. This project brings together many applications related to compliance with well-specified programming constraints to sign files electronically. The research has the goal of signing PDF documents digitally using the hash-and-sign method. The SHA-256 function as the hash function and RSA as encryption algorithm, accompanied with PKI certificates type X.509. The link between the applications and the certificates is the cryptographic API PKCS #11. This humbled work got dedicated to smartphones and computers. The use of WebView in all applications provides a taste of consistency. This article presents several processes carried out to accomplish this work.

scholarly journals PENERAPAN METODE ISYARAT TANGAN DALAM PEMBELAJARAN MENGHAFAL DAN MENGARTIKAN AL-QUR’AN

2020 ◽  
Vol 2 (1) ◽  
pp. 60-73
Author(s):  
Rahmiy Kurniasary ◽  
Ismail Sukardi ◽  
Ahmad Syarifuddin
Keyword(s):  
Hand Gesture ◽  
Hand Gestures ◽  
Internal Factors ◽  
Student Activities ◽  
Inhibiting Factor ◽  
Teacher Skills ◽  
Three Stages ◽  
Time Required ◽  
Sign Method ◽  
Descriptive Qualitative

Hand gesture method including requires high memorization ability, some students are not active and focus in synchronizing the pronunciation of lafadz verses and doing hand gestures in learning to memorize and interpret the Qur'an. The purpose of this study was to determine the application of the method of hand gesture in learning to memorize and interpret the Qur'an of students inX garade in Madrasah Aliyah Negeri 1 Prabumulih. The research method used is descriptive qualitative analysis that discusses the application of the method of hand gesture in learning to memorize and interpret the Qur'an of students inX grade in Madrasah Aliyah Negeri 1 Prabumulih. The type of approach used descriptive qualitative with data collection techniques through observation, interviews, documentation and triangulation. Analysis of data qualitatively through three stages, namely data reduction, data presentation and conclusion stages. The results of research conducted by researchers are, first, the steps in the application of hand sign method by the teacher of Al-Qur'an Hadith in X.IPA3 includes teacher activities, namely the teacher explains the material and gives examples of verses to be memorized and interpreted using method of hand gestures on learning video shows on the projector. Student activities, namely students apply the method of hand gesture to the verse that has been taught. Second, supporting factors in the application of hand gesture methods in the form of internal factors, namely from the level of willingness and ability to memorize, external namely in terms of the use of media, teacher skills and a pleasant learning atmosphere. Third, the inhibiting factor in the application of the hand gesture method is the time required by each student, the level of student willingness, skills in making hand gestures and synchronization between the pronunciation of lafadz with hand movements.

Enhancing adversarial attack transferability with multi-scale feature attack

2020 ◽  
pp. 2050076
Author(s):  
Caixia Sun ◽  
Lian Zou ◽  
Cien Fan ◽  
Yu Shi ◽  
Yifeng Liu
Keyword(s):  
Internal Representation ◽  
Feature Space ◽  
Source Model ◽  
Black Box ◽  
Space Representation ◽  
Scale Feature ◽  
Multi Scale ◽  
Adversarial Examples ◽  
Fast Gradient ◽  
Sign Method

Deep neural networks are vulnerable to adversarial examples, which can fool models by adding carefully designed perturbations. An intriguing phenomenon is that adversarial examples often exhibit transferability, thus making black-box attacks effective in real-world applications. However, the adversarial examples generated by existing methods typically overfit the structure and feature representation of the source model, resulting in a low success rate in a black-box manner. To address this issue, we propose the multi-scale feature attack to boost attack transferability, which adjusts the internal feature space representation of the adversarial image to get far to the internal representation of the original image. We show that we can select a low-level layer and a high-level layer of the source model to conduct the perturbations, and the crafted adversarial examples are confused with original images, not just in the class but also in the feature space representations. To further improve the transferability of adversarial examples, we apply reverse cross-entropy loss to reduce the overfitting further and show that it is effective for attacking adversarially trained models with strong defensive ability. Extensive experiments show that the proposed methods consistently outperform the iterative fast gradient sign method (IFGSM) and momentum iterative fast gradient sign method (MIFGSM) under the challenging black-box setting.

The Critical Velocity for an Infinite Cylindrical Shell under a Moving Load

2013 ◽  
Vol 441 ◽  
pp. 461-464
Author(s):  
Jiu Dan Zhang ◽  
Bin Zhen ◽  
Xiang Li
Keyword(s):  
Cylindrical Shell ◽  
Longitudinal Wave ◽  
Critical Velocity ◽  
Wave Velocity ◽  
Moving Load ◽  
Longitudinal Wave Velocity ◽  
Real Roots ◽  
Infinite Cylindrical Shell ◽  
Critical Velocities ◽  
Sign Method

The critical velocity for an infinite cylindrical shell subjected a moving load with a constant velocity is analyzed in this paper. It is found that the critical velocity problem can be translated into a distribution of the real roots of a quadruplicate equation, which can be solved by using Descartes sign method and complete discrimination system for polynomials. Our research shows that the number of the critical velocities for an infinite cylindrical shell always is even number. Furthermore the longitudinal wave velocity is not one critical velocity for the shell. Our results are different from the conclusion drawn by other authors that there are three critical velocities in an infinite shell, and the longitudinal wave velocity is the maximum critical velocity. Then further studies are needed to clarify these questions.

scholarly journals Random Transformation of image brightness for adversarial attack

2021 ◽  
pp. 1-12
Author(s):  
Bo Yang ◽  
Kaiyong Xu ◽  
Hengjun Wang ◽  
Hengwei Zhang
Keyword(s):  
Success Rate ◽  
Data Augmentation ◽  
Black Box ◽  
Image Brightness ◽  
Gradient Based ◽  
Adversarial Examples ◽  
Random Transformation ◽  
Fast Gradient ◽  
Adversarial Example ◽  
Sign Method

Deep neural networks (DNNs) are vulnerable to adversarial examples, which are crafted by adding small, human-imperceptible perturbations to the original images, but make the model output inaccurate predictions. Before DNNs are deployed, adversarial attacks can thus be an important method to evaluate and select robust models in safety-critical applications. However, under the challenging black-box setting, the attack success rate, i.e., the transferability of adversarial examples, still needs to be improved. Based on image augmentation methods, this paper found that random transformation of image brightness can eliminate overfitting in the generation of adversarial examples and improve their transferability. In light of this phenomenon, this paper proposes an adversarial example generation method, which can be integrated with Fast Gradient Sign Method (FGSM)-related methods to build a more robust gradient-based attack and to generate adversarial examples with better transferability. Extensive experiments on the ImageNet dataset have demonstrated the effectiveness of the aforementioned method. Whether on normally or adversarially trained networks, our method has a higher success rate for black-box attacks than other attack methods based on data augmentation. It is hoped that this method can help evaluate and improve the robustness of models.

Ictal Blinking: Reappraisal of the Lateralization and Localization Value in Focal Seizures

2022 ◽  
pp. 155005942110708
Author(s):  
Ayse Nur Ozdag Acarli ◽  
Ayse Deniz Elmali ◽  
Nermin Gorkem Sirin ◽  
Betul Baykan ◽  
Nerses Bebek
Keyword(s):  
The Other ◽  
Eeg Monitoring ◽  
Seizure Onset ◽  
Future Studies ◽  
Focal Seizures ◽  
Scalp Eeg ◽  
Ictal Activity ◽  
Generalized Epilepsy ◽  
Video Eeg ◽  
Sign Method

Introduction. Although ictal blinking is significantly more frequent in generalized epilepsy, it has been reported as a rare but useful lateralizing sign in focal seizures when it is not associated with facial clonic twitching. This study aimed to raise awareness of eye blinking as a semiological lateralizing sign. Method. Our database over an 11-year period reviewed retrospectively to assess patients who had ictal blinking associated with focal seizures. Results. Among 632 patients, 14 (2.2%), who had 3 to 13 (7 ± 3) seizures during video-EEG monitoring, were included. Twenty-five percent of all 92 seizures displayed ictal blinking and each patient had one to five seizures with ictal blinking. Ictal blinking was unilateral in 17%, asymmetrical in 22% and symmetrical in 61%. The blinking appeared with a mean latency of 6.3 s (range 0-39) after the clinical seizure-onset, localized most often to fronto-temporal, then in frontal or occipital regions. Blinking was ipsilateral to ictal scalp EEG lateralization side in 83% (5/6) of the patients with unilateral/asymmetrical blinking. The exact lateralization and localization of ictal activity could not have been determined via EEG in most of the patients with symmetrical blinking, remarkably. Conclusions. Unilateral/asymmetrical blinking is one of the early components of the seizures and appears as a useful lateralizing sign, often associated with fronto-temporal seizure-onset. Symmetrical blinking, on the other hand, did not seem to be valuable in lateralization and localization of focal seizures. Future studies using invasive recordings and periocular electrodes are needed to evaluate the value of blinking in lateralization and localization.

Sign in / Sign up

Export Citation Format

Share Document