# US 2025/0156532 A1 · Anomaly detection

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2025/0156532 A1
- Document ID: US20250156532A1
- Status: Published application
- Kind: A1
- Application number: 18/838,158
- Filed: 2023-02-07
- Published: 2025-05-15
- Earliest claimed: 2022-02-16
- Assignees: British Telecommunications public limited company
- Applicants: British Telecommunications Public Limited Company
- Inventors: Sadiq SANI
- Web page: https://patentgrep.com/patent/US20250156532A1

## CPC

- G06F 21/552: {involving long-term monitoring or reporting}
- H04L 63/1408: {by monitoring network traffic (monitoring network traffic per se H04L43/00)}
- G06N 3/0455: Auto-encoder networks; Encoder-decoder networks
- G06N 3/08: Learning methods
- G06N 3/088: Non-supervised learning, e.g. competitive learning
- G06N 3/09: Supervised learning
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS

## Abstract

A computer implemented method for detecting anomalous behaviour within a system is provided. The method generates an autoencoder for detecting anomalous behaviour within the system. The method also generates a classifier for predicting a classification of behaviour within the system. An input to the classifier comprises an output from one or more internal layers of the autoencoder. The method jointly trains the autoencoder and the classifier using a set of training data comprising a plurality of sample inputs that represent normal behaviour within the system and a plurality of sample inputs that represent anomalous behaviour within the system. The training is based on an output from a joint loss function that is configured to combine any reconstruction loss from the autoencoder with any prediction error from the classifier. The joint loss function is further configured to negate any reconstruction loss of the autoencoder in response to the sample input representing anomalous behaviour within the system. The trained autoencoder is used to classify a behaviour of the system as being normal or anomalous.

## Claims

### Claim 1

A computer implemented method for detecting anomalous behaviour within a system, the method comprising:
generating an autoencoder for detecting anomalous behaviour within the system;
generating a classifier for predicting a classification of behaviour within the system, wherein an input to the classifier comprises an output from one or more internal layers of the autoencoder;
jointly training the autoencoder and the classifier using a set of training data comprising a plurality of sample inputs that represent normal behaviour within the system and a plurality of sample inputs that represent anomalous behaviour within the system, the training being based on an output from a joint loss function that is configured to combine any reconstruction loss from the autoencoder with any prediction error from the classifier, wherein the joint loss function is further configured to negate any reconstruction loss of the autoencoder in response to the sample input representing anomalous behaviour within the system; and
using the trained autoencoder to classify a behaviour of the system as being normal or anomalous.

### Claim 2 (depends on 1)

The method of claim 1, wherein the autoencoder comprises a plurality of internal layers.

### Claim 3 (depends on 1)

The method of claim 1, wherein the input to the classifier comprises the output from an internal layer of the autoencoder that immediately precedes the output layer.

### Claim 4 (depends on 1)

The method of claim 1, wherein the input to the classifier consists of the output from a single internal layer of the autoencoder.

### Claim 5 (depends on 1)

The method of claim 1, wherein system is a computer network and the behaviour to be classified is an interaction between computer systems within the computer network.

### Claim 6 (depends on 5)

The method of claim 5, wherein a classification of the behaviour as anomalous indicates an attack on the computer network.

### Claim 7 (depends on 1)

A computer system comprising a processor and a memory storing computer program code for performing the method of claim 1.

### Claim 8 (depends on 1)

A computer program which, when executed by one or more processors, is arranged to carry out a method according to claim 1.
