Bit-level data generation and artificial intelligence techniques and architectures for data protection
AssigneeQuantum Star Technologies Inc.
InventorGarrett Thomas OETKEN
Life of this patent
Estimating the term…
Abstract
Techniques and architectures for representing data with one or more n-dimensional representations and/or using one or more models to identify threats associated with the one or more n-dimensional representations are described herein. For example, the techniques and architectures may determine one or more coordinates for one or more points based on one or more sets of bits in the data and generate an n-dimensional representation for the data based on the one or more points. The techniques and architectures may evaluate the n-dimensional representation with one or more machine-trained models to detect a threat associated with the data, such as malware or another threat.
Claims20
A method of detecting malware, the method comprising:
- receiving, by a computing device, data from a data store;
- identifying, by the computing device, at least a first group of bits in the data and a second group of bits in the data;
- representing, by the computing device, a first set of bits in the first group of bits as a first coordinate for a first point and a second set of bits in the first group of bits as a second coordinate for the first point;
- representing, by the computing device, a first set of bits in the second group of bits as a first coordinate for a second point and a second set of bits in the second group of bits as a second coordinate for the second point;
- generating, by the computing device, an n-dimensional representation for the data based at least in part on the first point and the second point;
- processing the n-dimensional representation using a model that has been trained using machine learning;
- and
- determining a malware rating for the data based at least in part on the processing, the malware rating indicating a likelihood that the data is associated with malware.
A system comprising:
- control circuitry;
- and
- memory communicatively coupled to the control circuitry and storing executable instructions that, when executed by the control circuitry, cause the control circuitry to perform operations comprising:
- obtaining data;
- determining a first coordinate for a first point based at least in part on a first set of bits in the data and determining a second coordinate for the first point based at least in part on a second set of bits in the data that is adjacent to the first set of bits;
- determining a first coordinate for a second point based at least in part on a third set of bits in the data and determining a second coordinate for the second point based at least in part on a fourth set of bits in the data that is adjacent to the third set of bits;
- generating an n-dimensional representation for the data based at least in part on the first point and the second point;
- and
- causing the n-dimensional representation to be processed with a machine-trained model that is configured to detect malware.
One or more non-transitory computer-readable media storing computer-executable instructions that, when executed, instruct one or more processors to perform operations comprising:
- obtaining data;
- determining a first coordinate for a first point based at least in part on a first set of bits in the data and determining a second coordinate for the first point based at least in part on a second set of bits in the data that is adjacent to the first set of bits;
- determining a first coordinate for a second point based at least in part on a third set of bits in the data and determining a second coordinate for the second point based at least in part on a fourth set of bits in the data that is adjacent to the third set of bits;
- generating an n-dimensional representation for the data based at least in part on the first and second coordinates for the first point and the first and second coordinates for the second point;
- and
- causing the n-dimensional representation to be processed with a machine-trained model that is configured to detect a threat.
Dependencies are inferred from the claim text.