18108727. METHOD AND APPARATUS WITH NEURAL NETWORK TRAINING simplified abstract (Samsung Electronics Co., Ltd.)

METHOD AND APPARATUS WITH NEURAL NETWORK TRAINING

Organization Name

Samsung Electronics Co., Ltd.

Inventor(s)

Nam Soo Kim of Seoul (KR)

Jiwon Yoon of Seoul (KR)

METHOD AND APPARATUS WITH NEURAL NETWORK TRAINING - A simplified explanation of the abstract

This abstract first appeared for US patent application 18108727 titled 'METHOD AND APPARATUS WITH NEURAL NETWORK TRAINING

Simplified Explanation

The patent application describes a method and apparatus for training a neural network using latent target data.

• Predict first latent target data based on source data and corresponding target data.
• Predict second latent target data based on source data and constant data.
• Train the neural network model using the first latent target data and predicted second latent target data.
• The first latent target data and target data have a many-to-one relationship.

---

1. 1. 1. Potential Applications

• Machine learning
• Artificial intelligence
• Data analysis

1. 1. 1. Problems Solved

• Improving accuracy of neural network models
• Enhancing predictive capabilities
• Handling complex relationships between data

1. 1. 1. Benefits

• Increased efficiency in training neural networks
• Improved performance in predicting target data
• Enhanced capabilities in handling diverse datasets

Original Abstract Submitted

Provided is a method and an apparatus with neural network (NN) training. A method of operating a neural network model includes predicting first latent target data based on source data and based on target data corresponding to the source data, predicting second latent target data based on the source data and based on constant data, and training the NN model based on the first latent target data and the predicted second latent target data; the first latent target data and the target data have a many-to-one relationship.