17957889. METHOD, APPARATUS, AND DEVICE FOR OBTAINING ARTIFICIAL INTELLIGENCE MODEL, AND STORAGE MEDIUM simplified abstract (HUAWEI TECHNOLOGIES CO., LTD.)

From WikiPatents
Jump to navigation Jump to search

METHOD, APPARATUS, AND DEVICE FOR OBTAINING ARTIFICIAL INTELLIGENCE MODEL, AND STORAGE MEDIUM

Organization Name

HUAWEI TECHNOLOGIES CO., LTD.

Inventor(s)

Xiaoyun Si of Nanjing (CN)

Xinyu Hu of Nanjing (CN)

Li Xue of Nanjing (CN)

Liang Zhang of Nanjing (CN)

Fuxing Chen of Boulogne Billancourt (FR)

METHOD, APPARATUS, AND DEVICE FOR OBTAINING ARTIFICIAL INTELLIGENCE MODEL, AND STORAGE MEDIUM - A simplified explanation of the abstract

This abstract first appeared for US patent application 17957889 titled 'METHOD, APPARATUS, AND DEVICE FOR OBTAINING ARTIFICIAL INTELLIGENCE MODEL, AND STORAGE MEDIUM

Simplified Explanation

The patent application describes a method, apparatus, and device for obtaining an artificial intelligence (AI) model. The client receives an initial AI model from a service end, which consists of multiple neurons. During subsequent rounds of training, the client selects a target neuron from the initial model and trains it using local data. The client then returns the parameter data of the target neuron to the service end, which uses it to obtain a converged target AI model.

  • The client receives an initial AI model from a service end.
  • The initial AI model consists of multiple neurons.
  • During subsequent rounds of training, the client selects a target neuron.
  • The target neuron is trained using local data.
  • The client returns the parameter data of the target neuron to the service end.
  • The service end uses the parameter data to obtain a converged target AI model.

Potential Applications

  • This technology can be applied in various fields that utilize AI models, such as image recognition, natural language processing, and autonomous vehicles.
  • It can be used in distributed AI systems where multiple clients contribute to the training process.

Problems Solved

  • The method allows for distributed training of AI models, where clients can train specific neurons based on their local data.
  • It enables the convergence of a target AI model by combining the trained neurons from multiple clients.

Benefits

  • The method allows for efficient training of AI models by utilizing local data from multiple clients.
  • It enables the improvement of AI models through collaborative training from different sources.
  • The approach reduces the need for centralized data storage and processing, enhancing privacy and scalability.


Original Abstract Submitted

A method, an apparatus, and a device for obtaining an artificial intelligence model, and a storage medium are provided. A client receives a first artificial intelligence AI model sent by a service end (). The first AI model includes a plurality of neurons. The client determines, from the plurality of neurons, a target neuron participating in a current round of training, where the current round of training is a non-first round of training, and a quantity of target neurons is less than a total quantity of the plurality of neurons (). The client trains the target neuron based on local data (). The client returns parameter data corresponding to the target neuron to the service end (). The parameter data corresponding to the target neuron is used by the service end to obtain a converged target AI model.

Retrieved from "http://wikipatents.org/index.php?title=17957889._METHOD,_APPARATUS,_AND_DEVICE_FOR_OBTAINING_ARTIFICIAL_INTELLIGENCE_MODEL,_AND_STORAGE_MEDIUM_simplified_abstract_(HUAWEI_TECHNOLOGIES_CO.,_LTD.)&oldid=20048"