Invention Grant
- Patent Title: Gaussian autoencoder detection of network flow anomalies
-
Application No.: US16584552Application Date: 2019-09-26
-
Publication No.: US11297096B2Publication Date: 2022-04-05
- Inventor: Petr Gronát , Mikulá{hacek over (s)} Zelinka
- Applicant: Avast Software s.r.o.
- Applicant Address: CZ Prague
- Assignee: Avast Software s.r.o.
- Current Assignee: Avast Software s.r.o.
- Current Assignee Address: CZ Prague
- Agency: Avek IP, LLC
- Agent William B. Kircher
- Main IPC: H04L29/06
- IPC: H04L29/06 ; G06N3/08 ; G06K9/62

Abstract:
A method of identifying malicious activity in a computer data sequence includes providing provided the computer data sequence to a network configured to convert the computer data sequence from a high-dimensional space to a low-dimensional space, and processing the computer data sequence in the low-dimensional space to generate an approximately Gaussian distribution. The processed computer data sequence converted to the low dimensional space is evaluated relative to the approximately Gaussian distribution to determine whether the computer data sequence is likely malicious or likely benign, and an output is provided indicating whether the computer data sequence is likely malicious or likely benign.
Information query