Writing
Deep dives into machine learning theory, PyTorch experiments, computer vision, NLP, and AI in cybersecurity.
7 articles
In this tutorial, we apply Graph Convolutional Network (GCN) and Graph Attention Network (GAT) to detect fraudulent Bitcoin transactions on the Elliptic dataset, and compare their performances.
Most websites use JavaScript to make dynamic content, which makes it a valuable attack vector against browsers and other JS applications. Most malicious JS is obfuscated to hide its behavior. Here we explore static analysis approaches using NLP features and standard ML classifiers.
According to OWASP ratings, the problem of protection in the digital space continues to be extremely relevant. Here we consider actual challenges arising from the conjunction of machine learning and cybersecurity in areas such as IoT ecosystems, targeted APTs, fraud detection, ransomware, and adversarial attacks.
Phishing is a fraudulent activity that aims to steal user credentials, credit card and bank account information or to deploy malicious software on the victim's infrastructure. Here we discuss URL-based detection using standard ML approaches in combination with NLP features extracted from the URL.
In this tutorial we consider colorectal histology tissues classification using ResNet-50 architecture and PyTorch framework. We train a CNN from scratch on 8 tissue classes and achieve 92% accuracy on the test set.
This tutorial contains brief overview of statistical and machine learning methods for time series forecasting, experiments and comparative analysis of Long sh
In this article, we will consider key concepts of discriminative and generative models, differences between them. In the light of these concepts, we will desc