Darktrace AI: Combining Unsupervised and Supervised Machine Learning

A New Age of Cyber-Threat

  Supervised learning works by using previously-classified data, from which the machine learns the classification system. For scenarios where behaviors are well understood and classifications are easy to determine, the output of such systems can be highly accurate.
  For example, state-of-the-art image classification systems are outperforming humans in some cases. Indeed, what makes supervised machine learning so powerful is itsability to learn to deal with the errors and noise of the real world, through a statistical
approach.

Core Principles of Darktrace’s Machine Learning:

• Learns ‘on the job’ – it does not depend upon knowledge of previous attacks.
• Thrives on complexity and diversity of modern businesses.
• Constantly revises assumptions about behavior, using probabilistic mathematics.
• Always up to date, and not reliant on human input.