Adversarial Attacks on Deep Learning Systems

Deep learning is at the heart of the current rise of artificial intelligence. In the field of Computer Vision, it has become the workhorse for applications ranging from self-driving cars to surveillance and security. Whereas deep neural networks have demonstrated phenomenal success (often beyond human capabilities) in solving complex problems, recent studies show that they are vulnerable to adversarial attacks in the form of subtle perturbations to inputs that lead a model to predict incorrect outputs. For images, such perturbations are often too small to be perceptible, yet they completely fool the deep learning models. Adversarial attacks pose a serious threat to the success of deep learning in practice. This fact has recently lead to a large influx of contributions in this direction. This article presents a survey on adversarial attacks on deep learning in Computer Vision. We review the works that design adversarial attacks, analyze the existence of such attacks and propose defenses against them

أنظمة تركيب الكلام

يعتبر تركيب الكلام بشكل آلي من أهم المسائل المطروحة في الذكاء الصنعي منذ بداياته، حيث تتزايد الحاجة لأنظمة تركيب كلام مستقرة بحيث تنتج خطاباً ذكياً مع كيفية الكلام الطبيعي. تركيب الكلام speech synthesis يعتبر الأهم في تصنيع آلات متحدثة قادرة على التواصل مع الإنسان بطريقة فعالة وبسيطة، سواء في أنظمة الحوار الآلي Auto dialogue systems، توليد آلي للكتب الصوتية Auto generating of Audio books، متحدث مساعد لذوي الاحتياجات الخاصة different able humans، وغير ذلك من الأنظمة التي تحتاجها البشرية اعتماداً على معالجة اللغات الطبيعيةNatural language processing وفهم الكلام Language Understanding في سبيل خدمة التواصل بين الآلة والإنسان Human-computer Interaction.

Recognition of Hand Written Arabic Names Using Deep Learning

Designing Computerized Systems which posses reading and hearing faculties is an active research area for more than four decades. Many methods and algorithms have been suggested by researches for this purpose as part of pattern recognition research. Recently, more research work has been devoted to the holist approach the recognition system recognizes a complete word as one object without going through the long and erroneous character segmentation process. In this paper, a convolutional neural network has been designed to recognize the popular Arabic names holistically. SUSt ARG names data set has been used to test the network performance (collected and compiled by pattern recognition research in Sudan University of Science and Technology-SUSt). Selecting an appropriate deep learning toolbox, after five stages of training, the network was able to recognize all the names and 100%

Design a CNN network for vehicle classification

The project aims primarily to employ the benefits of artificial intelligence, specifically the characteristics of programming a neuronal network where neuronal networks, in turn, are networks that are interested in training and learning from error, and employing this error to achieve optimal results.Convolution NeuralNetworks(CNN)in particular are one of the most important neuronal networks that address classification problems and issues. Thus, this project is to design a convolution neuronal network that classifies vehicles into several types where we will design the network and train them on the database as the database includes pictures of several types of vehicles The network will classify each Image to its type, after adjusting the images, making the appropriate changes, turning them gray, and discovering the edges and lines.After the images are ready, the training process will begin, and after the training process is finished, we will produce classification results, and then we will test with a new set of images.One of the most important applications of this project is to abide by the paving places of cars, trucks, and vehicles in general, as if a picture was entered as a car for the car sample, which is a truck, for example, this will give an error where the network will discover this by examining and classifying it. As a truck, we discover that there is a violation of the paving laws

التصنيف التلقائي للهياكل الجيولوجية باستخدام الرؤية الحاسوبية ونموذج التعلم العميق

دراسة الهياكل الجيولوجية المكشوفة على سطح الأرض ذات أهمية كبيرة بشكل عام وخصوصا في التصميم الهندسي والبناء. في هذا البحث ، استخدمنا 2206 صورة مع 12 ملصق للتعرف على الهياكل الجيولوجية بناءً على نموذج Inception-v3. تم اعتماد الصور ذات التدرج الرمادي واللون في النموذج. كما تم بناء نموذج الشبكة العصبية التلافيفية (CNN) وتم تطبيق خوارزمية أقرب جار (KNN) والشبكة العصبية الاصطناعية (ANN) وتعزيز التدرج الشديد (XGBoost) في تصنيف الهياكل الجيولوجية بناءً على الميزات المستخرجة من مكتبة رؤية الكمبيوتر مفتوحة المصدر (OpenCV). أخيرًا ، تمت مقارنة أداء الطرق الخمس وأظهرت النتائج أن أداء KNN و ANN و XGBoost كان ضعيفًا وبدقة أقل من 40.0٪. أما CNN فعد عانت من فرط التدريب Overfitting. كان للنموذج الذي تم تدريبه باستخدام التعلم بالنقل تأثير كبير على مجموعة بيانات صغيرة من صور التركيب الجيولوجي. وأفضل نموذجين وصلوا إلى دقة 83.3٪ و 90.0٪ على التوالي. هذا يدل على أن النسيج هو السمة الرئيسية في هذا البحث. يمكن أن يستخرج التعلم القائم على نموذج التعلم العميق ميزات بيانات البنية الجيولوجية الصغيرة بشكل فعال ، وهو قوي في تصنيف صور الهيكل الجيولوجي.

A Bag of Tricks for Dialogue Summarization

Dialogue summarization comes with its own peculiar challenges as opposed to news or scientific articles summarization. In this work, we explore four different challenges of the task: handling and differentiating parts of the dialogue belonging to multiple speakers, negation understanding, reasoning about the situation, and informal language understanding. Using a pretrained sequence-to-sequence language model, we explore speaker name substitution, negation scope highlighting, multi-task learning with relevant tasks, and pretraining on in-domain data. Our experiments show that our proposed techniques indeed improve summarization performance, outperforming strong baselines.

Effect of Changing CNN Classifier Parameters on EEG Signals Recognition Ratio

Brain Computer Interface (BCI), especially systems for recognizing brain signals using deep learning after characterizing these signals as EEG (Electroencephalography), is one of the important research topics that arouse the interest of many researchers currently. Convolutional Neural Nets (CNN) is one of the most important deep learning classifiers used in this recognition process, but the parameters of this classifier have not yet been precisely defined so that it gives the highest recognition rate and the lowest possible training and recognition time. This research proposes a system for recognizing EEG signals using the CNN network, while studying the effect of changing the parameters of this network on the recognition rate, training time, and recognition time of brain signals, as a result the proposed recognition system was achieved 76.38 % recognition rate, And the reduction of classifier training time (3 seconds) by using Common Spatial Pattern (CSP) in the preprocessing of IV2b dataset, and a recognition rate of 76.533% was reached by adding a layer to the proposed classifier.