14 Posts

Brad Harris

Security Researcher, IBM X-Force

    Brad has worked in the network and computer security field in both the public and private sectors. He has done everything from conducting penetration tests to reverse engineering to applied research. Currently he is a researcher for IBM's X-Force.

    Written By Brad Harris

    Now That You Have a Machine Learning Model, It’s Time to Evaluate Your Security Classifier

    Now that you've identified an AI solution and selected a suitable algorithm for your machine learning model, you're ready to measure the effectiveness of your security classifier.

    Generative Adversarial Networks and Cybersecurity: Part 2

    Researchers have shown how generative adversarial networks (GANs) can be applied to cybersecurity tasks such as cracking passwords and identifying hidden data in high-quality images.

    Generative Adversarial Networks and Cybersecurity: Part 1

    Generative adversarial networks are neural networks that compete in a game in which a generator attempts to fool a discriminator with examples that look similar to a training set.

    With AI2, Machine Learning and Analysts Come Together to Impress, Part 3: The Experiment

    Two researchers performed an experiment based on billions of log lines that demonstrated the importance of domain expertise in machine learning analysis.

    With AI2, Machine Learning and Analysts Come Together to Impress, Part 2: The Algorithms

    AI2 uses an "analyst-in-the-loop" system to improve itself and a "human-in-the-loop" system to create examples to be used in iterative training algorithms.

    With AI2, Machine Learning and Analysts Come Together to Impress, Part 1: An Introduction

    Machine learning systems like A12 are designed to augment human analysis with cognitive intelligence, enabling IT professionals to reduce false positives.

    Security Challenges in the Cloud Environment

    Many enterprises are turning to the cloud for better availability and storage, but there are a number of threats that can put those environments at risk.

    Factorization Machines: A New Way of Looking at Machine Learning

    Here's a look at Steffen Rendle's theory on factorization machines — including how they could be used to assist data scientists in the future.

    Side-Channel Attacks Against Multicore Processors in Cross-VM Scenarios: Part III

    In the final installment of this series, we focus on S$A attacks as well as some ways your organization can prevent side-channel attacks on its VMs.

    Side-Channel Attacks Against Multicore Processors in Cross-VM Scenarios: Part II

    Discussion of two side-channel attacks meant to retrieve sensitive information from a virtual machine (VM) on the same physical processor package.

    Co-Written By Brad Harris

    Machine Learning Algorithms Are Not One-Size-Fits-All

    Before you can choose the right machine learning algorithms to serve your business' needs, you must understand the type of problem you're trying to solve and the type of training data you'll need.