FireEye brings machine learning to endpoint security solution

Security company FireEye has announced the addition of MalwareGuard – a new machine learning-based detection and prevention engine – to its Endpoint Security solution.

MalwareGuard is designed to help detect and block cyber threats including never-before-seen threats to provide customers an added level of protection to stop attacks and protect customer information, sensitive data, and intellectual property.

MalwareGuard is integrated into the FireEye Endpoint Security agent and is available now for current customers at no additional cost, and via a free trial for other organisations interested in upgrading their endpoint defences.

FireEye MalwareGuard is the result of a two-year research project from FireEye data scientists and testing in real-world incident responses.

The MalwareGuard model is trained using machine learning techniques to enable it to make intelligent malware classifications on its own and without human involvement.

The machine learning model is trained with both public and private data sources, including data gathered from over 15 million endpoint agents, attack analyses based on more than one million hours spent responding to attacks to date, over 200,000 consulting hours every year and adversarial intelligence collected from a global network of analysts that speak 32 languages.

With the addition of MalwareGuard, FireEye Endpoint Security agent now includes four integrated engines: machine learning (MalwareGuard), behaviour-based (ExploitGuard), signature-based (Malware Protection) and intelligence-based (IOC), to provide a layered defence designed to protect customers from known and unknown threats.

These engines are updated with FireEye's threat intelligence and designed to keep pace with evolving threats.

In addition to prevention engines, FireEye Endpoint Security includes investigation, detection and response (EDR) capabilities designed to enable organisations to rapidly investigate and respond to attacks on the endpoint.

This is included in a lightweight agent and managed through the cloud, on-premises or a hybrid deployment. 

FireEye engineering senior vice president John Laliberte says, "Attackers are constantly innovating and outmanoeuvring legacy, signature-based technology.

"Reducing the window of time from discovery to analysis, and deployment of protection is critical to reducing risk in an enterprise.

"By combining our frontline knowledge of the adversaries with our in-house machine learning expertise, we can now better protect our customers against cyber threats including never-before-seen threats by automating the discovery, analysis, and deployment of protection through our endpoint solution.

In addition to the new machine learning capabilities, FireEye Endpoint Security now includes new management features that simplify the process of moving from alert to fix.

These include:

• Policy Manager: enhanced enterprise management makes it easy to enable varying levels of access, allowing administrators to balance the needs of security and performance
• Alert Workflow Update: provides the necessary context for organisations to rapidly respond to the alerts that matter
• Cloud Identity and Access Management: enables a higher level of authentication for cloud-based deployments