18262098. LEARNING APPARATUS, ANOMALY DETECTION APPARATUS, LEARNING METHOD, ANOMALY DETECTION METHOD, AND PROGRAM simplified abstract (NIPPON TELEGRAPH AND TELEPHONE CORPORATION)
Contents
- 1 LEARNING APPARATUS, ANOMALY DETECTION APPARATUS, LEARNING METHOD, ANOMALY DETECTION METHOD, AND PROGRAM
- 1.1 Organization Name
- 1.2 Inventor(s)
- 1.3 LEARNING APPARATUS, ANOMALY DETECTION APPARATUS, LEARNING METHOD, ANOMALY DETECTION METHOD, AND PROGRAM - A simplified explanation of the abstract
- 1.4 Simplified Explanation
- 1.5 Potential Applications
- 1.6 Problems Solved
- 1.7 Benefits
- 1.8 Potential Commercial Applications
- 1.9 Possible Prior Art
- 1.10 Original Abstract Submitted
LEARNING APPARATUS, ANOMALY DETECTION APPARATUS, LEARNING METHOD, ANOMALY DETECTION METHOD, AND PROGRAM
Organization Name
NIPPON TELEGRAPH AND TELEPHONE CORPORATION
Inventor(s)
LEARNING APPARATUS, ANOMALY DETECTION APPARATUS, LEARNING METHOD, ANOMALY DETECTION METHOD, AND PROGRAM - A simplified explanation of the abstract
This abstract first appeared for US patent application 18262098 titled 'LEARNING APPARATUS, ANOMALY DETECTION APPARATUS, LEARNING METHOD, ANOMALY DETECTION METHOD, AND PROGRAM
Simplified Explanation
The learning apparatus described in the abstract is designed to train a model using data collections from two different systems, with the goal of distinguishing between features of the target domain and the source domain.
- The learning apparatus includes an input unit for normal data collections from the target and source domains.
- A learning unit trains a model with two autoencoders and a discriminator to distinguish features between the two domains.
- The first autoencoder processes data from the target domain, while the second autoencoder processes data from the source domain.
- The discriminator outputs the probability that output data represents a feature from either domain.
Potential Applications
This technology could be applied in various fields such as anomaly detection, fraud detection, and cybersecurity to identify and distinguish features between different domains.
Problems Solved
This technology addresses the challenge of distinguishing features between different systems or domains, which can be useful in detecting anomalies or identifying patterns in data.
Benefits
The learning apparatus can improve the accuracy and efficiency of feature detection and classification in complex datasets from multiple domains.
Potential Commercial Applications
One potential commercial application of this technology could be in the development of advanced security systems for detecting fraudulent activities in financial transactions.
Possible Prior Art
Prior art in the field of machine learning and anomaly detection may include similar approaches to distinguishing features between different domains using autoencoders and discriminators.
What are the limitations of this technology in real-world applications?
One limitation of this technology in real-world applications could be the computational resources required to train and deploy the model, especially for large-scale datasets.
How does this technology compare to existing methods for feature detection and classification?
This technology offers a novel approach to distinguishing features between different domains using autoencoders and discriminators, which may provide more accurate and reliable results compared to traditional methods.
Original Abstract Submitted
A learning apparatus according to one embodiment includes an input unit configured to input a normal data collection for a first system that is a target domain and to input a normal data collection for a second system that is a source domain. The learning apparatus includes a learning unit configured to train a model that includes a first autoencoder configured to input normal data for the target domain, based on the normal data collection for the first system and the normal data collection for the second system. The model includes a second autoencoder configured to input normal data for the source domain, and includes a discriminator configured to output a probability that output data is data representing a feature for any one of the target domain and the source domain, while using, as an input, output data, output data of a first encoder included in the first autoencoder or a second encoder included in the second autoencoder.