Hiding Malware in ML Models

Interesting research: "EvilModel: Hiding Malware Inside of Neural Network Models". Abstract: Delivering malware covertly and detection-evadingly is critical to advanced malware campaigns. In this paper, we present a method that delivers malware covertly and detection-evadingly through neural...

In0ri - Defacement Detection With Deep Learning

In0ri is a defacement detection system utilizing a image-classification convolutional neural network. Introduction When monitoring a website, In0ri will periodically take a screenshot of the website then put it through a preprocessor that will resize the image down to 250x250px and numericalize t...

PT-2021-3272 · Nni · Nni

Name of the Vulnerable Software and Affected Versions: NNI versions affected versions not specified Description: The issue is related to incorrect code generation management in the common utils.py module of the Neural Network Intelligence NNI toolkit, which is used for automating design, neural...

Training Transformers for Cyber Security Tasks: A Case Study on Malicious URL Prediction

Highlights Perform a case study on using Transformer models to solve cyber security problems Train a Transformer model to detect malicious URLs under multiple training regimes Compare our model against other deep learning methods, and show it performs on-par with other top-scoring models Identify...

Machine learning classifiers trained via gradient descent are vulnerable to arbitrary misclassification attack

Overview Machine learning models trained using gradient descent can be forced to make arbitrary misclassifications by an attacker that can influence the items to be classified. The impact of a misclassification varies widely depending on the ML model's purpose and of what systems it is a part...

Eyeballer - Convolutional Neural Network For Analyzing Pentest Screenshots

Give those screenshots of yours a quick eyeballing. Eyeballer is meant for large-scope network penetration tests where you need to find "interesting" targets from a huge set of web-based hosts. Go ahead and use your favorite screenshotting tool like normal EyeWitness or GoWitness and then run the...

PYSEC-2019-107

nbla/logger.cpp in libnnabla.a in Sony Neural Network Libraries aka nnabla through v1.0.14 relies on the HOME environment variable, which might be untrusted...

Code injection

nbla/logger.cpp in libnnabla.a in Sony Neural Network Libraries aka nnabla through v1.0.14 relies on the HOME environment variable, which might be untrusted...

CVE-2019-10844

nbla/logger.cpp in libnnabla.a in Sony Neural Network Libraries aka nnabla through v1.0.14 relies on the HOME environment variable, which might be untrusted...

PYSEC-2019-107

nbla/logger.cpp in libnnabla.a in Sony Neural Network Libraries aka nnabla through v1.0.14 relies on the HOME environment variable, which might be untrusted...

CVE-2019-10844

nbla/logger.cpp in libnnabla.a in Sony Neural Network Libraries aka nnabla through v1.0.14 relies on the HOME environment variable, which might be untrusted...

PYSEC-2019-37

nbla/logger.cpp in libnnabla.a in Sony Neural Network Libraries aka nnabla through v1.0.14 relies on the HOME environment variable, which might be untrusted...

CVE-2019-10844

CVE-2019-10844 affects Sony Neural Network Libraries (nnabla) – nbla/logger.cpp in libnnabla.a up to v1.0.14. The root cause is that code relies on the HOME environment variable, which is untrusted, enabling potential influence on behavior via the user’s HOME value. Public references in Red Hat a...

CVE-2019-10844

nbla/logger.cpp in libnnabla.a in Sony Neural Network Libraries aka nnabla through v1.0.14 relies on the HOME environment variable, which might be untrusted...

Wallarm New Open Source Module and Kaggle Hackathon

A key element of any security solution, whether its a WAF, NGWAF, RASP or even a SIEM or a classic IDS, is the ability to correctly detect whether an incoming API request is malicious. The traditional way to do it is using signatures and regular expressions regex. Some sets of signatures are...

Detecting Fake Videos

This story nicely illustrates the arms race between technologies to create fake videos and technologies to detect fake videos: These fakes, while convincing if you watch a few seconds on a phone screen, aren't perfect yet. They contain tells, like creepily ever-open eyes, from flaws in their...

TensorFlow Dataset API for increasing training speed of neural networks

by M.Salnikov, Wallarm Research Wallarm AI engine is the heart of our security solution. Two key parameters of our AI engine efficiency are how fast neural networks can be train to reflect the updated training sets and how much compute power need to be dedicated to the training on the on-going...

Sit-down with Wallarm CTO, Alex Golovko

I have had a chance to pose a few questions to Alexander Golovko, one of the co-founders of Wallarm and our CTO. Here are Alex’s reflections on Wallarm and some technology trends. How did Wallarm get its start? Ivan Wallarm’s founder has involved me in various projects on and off since 2010. By...

The First Step-by-Step Guide for Implementing Neural Architecture Search with Reinforcement…

The First Step-by-Step Guide for Implementing Neural Architecture Search with Reinforcement Learning Using TensorFlow Our team is no stranger to various flavors of AI including deep learning DL. That’s why we’ve immediately noticed when Google came out with AutoML project, designed to make AI bui...

Apple iPhone X's Face ID Hacked (Unlocked) Using 3D-Printed Mask

Just a week after Apple released its brand new iPhone X on November 3, a team of hackers has claimed to successfully hack Apple's Face ID facial recognition technology with a mask that costs less than $150. Yes, Apple's "ultra-secure" Face ID security for the iPhone X is not as secure as the...