Method and apparatus for performing anomaly detection using neural network
AssigneesSamsung Electronics Co., Ltd., Postech Research and Business Development Foundation
InventorsHyunsoo Kim, Jaeyoon Sim, Jaehan Park, Hyunwoo Son, Sangjoon Kim
Life of this patent
Estimating the term…
Abstract
A method and apparatus for performing anomaly detection by using a neural network are provided. The apparatus is configured to extract input features of an input data signal, obtain output features of the neural network by processing the input features through the neural network, obtain an error based on the input features and the output features, and determine whether the input data signal indicates an abnormal signal based on the error and a threshold.
Claims13
A method of performing an anomaly detection using a neural network including layers of an encoder and a decoder, the method comprising:
- processing, by a processing circuitry, input features of an input data signal using the neural network such that output features of the neural network corresponding to an output of the decoder are obtained;
- determining, by the processing circuitry, whether the input data signal indicates an abnormal signal or a normal signal based on an error between the input features and the output features;
- and
- outputting, by the processing circuitry, information indicating that the abnormal signal is detected based on a determination that the input data signal indicates the abnormal signal,
- wherein the neural network is configured to perform unsupervised learning to differentiate the abnormal signal from the normal signal by updating weights of an input layer and an output layer of the neural network through online learning, when the input data signal indicates the normal signal, such that the error calculated as a difference between the input features and the output features decreases according to a gradient descent through the unsupervised learning.
An apparatus for performing anomaly detection by using a neural network including layers of an encoder and a decoder, the apparatus comprising:
- a memory storing at least one program;
- and
- processing circuitry configured to perform the anomaly detection by executing the at least one program,
- wherein the processing circuitry is configured to:
- process input features of an input data signal using the neural network such that output features of the neural network corresponding to an output of the decoder are obtained,
- determine whether the input data signal indicates an abnormal signal or a normal signal based on an error between the input features and the output features, and
- output, by the processing circuitry, information indicating that the abnormal signal is detected based on a determination that the input data signal indicates the abnormal signal,
- wherein the neural network is configured to perform unsupervised learning to differentiate the abnormal signal from the normal signal by updating weights of an input layer and an output layer of the neural network through online learning, when the input data signal indicates the normal signal, such that the error calculated as a difference between the input features and the output features decreases according to a gradient descent through the unsupervised learning.
Dependencies are inferred from the claim text.