17950449. ARTIFICIAL NEURAL NETWORK MODULE FOR PERFORMING ARTIFICIAL NEURAL NETWORK OPERATION ON PLURALITY OF SUBGRAPHS AND OPERATING METHOD THEREOF simplified abstract (Samsung Electronics Co., Ltd.)
Contents
- 1 ARTIFICIAL NEURAL NETWORK MODULE FOR PERFORMING ARTIFICIAL NEURAL NETWORK OPERATION ON PLURALITY OF SUBGRAPHS AND OPERATING METHOD THEREOF
- 1.1 Organization Name
- 1.2 Inventor(s)
- 1.3 ARTIFICIAL NEURAL NETWORK MODULE FOR PERFORMING ARTIFICIAL NEURAL NETWORK OPERATION ON PLURALITY OF SUBGRAPHS AND OPERATING METHOD THEREOF - A simplified explanation of the abstract
- 1.4 Simplified Explanation
- 1.5 Potential Applications
- 1.6 Problems Solved
- 1.7 Benefits
- 1.8 Original Abstract Submitted
ARTIFICIAL NEURAL NETWORK MODULE FOR PERFORMING ARTIFICIAL NEURAL NETWORK OPERATION ON PLURALITY OF SUBGRAPHS AND OPERATING METHOD THEREOF
Organization Name
Inventor(s)
Seungsoo Yang of Hwaseong-si (KR)
ARTIFICIAL NEURAL NETWORK MODULE FOR PERFORMING ARTIFICIAL NEURAL NETWORK OPERATION ON PLURALITY OF SUBGRAPHS AND OPERATING METHOD THEREOF - A simplified explanation of the abstract
This abstract first appeared for US patent application 17950449 titled 'ARTIFICIAL NEURAL NETWORK MODULE FOR PERFORMING ARTIFICIAL NEURAL NETWORK OPERATION ON PLURALITY OF SUBGRAPHS AND OPERATING METHOD THEREOF
Simplified Explanation
The abstract describes a method for operating an artificial neural network on multiple subgraphs. Here is a simplified explanation:
- The method involves generating a trigger for determining the resources needed for a specific subgraph within a neural network model.
- A control signal is then generated to allocate hardware resources to the target subgraph based on the trigger.
- The hardware allocation and resource settings are adjusted according to the control signal.
- Finally, an operation is performed on the target subgraph using the modified hardware and resource settings.
Potential Applications
This technology has potential applications in various fields, including:
- Machine learning: The method can be used to optimize the performance of artificial neural networks, improving their accuracy and efficiency.
- Computer vision: By efficiently allocating resources to different subgraphs, this method can enhance the processing of visual data, enabling better object recognition and image analysis.
- Natural language processing: The technique can be applied to neural networks used for language processing tasks, such as speech recognition or sentiment analysis, improving their overall performance.
Problems Solved
The method addresses several problems in neural network operations, including:
- Resource allocation: By dynamically allocating hardware resources to different subgraphs, the method optimizes the utilization of available resources, leading to improved performance and efficiency.
- Scalability: The approach allows for the operation of large neural network models with multiple subgraphs, ensuring efficient processing and avoiding bottlenecks.
- Flexibility: The method enables the adaptation of hardware and resource settings based on the specific requirements of each subgraph, allowing for better customization and performance optimization.
Benefits
The use of this technology offers several benefits, including:
- Improved performance: By allocating resources based on the specific needs of each subgraph, the method enhances the overall performance of the neural network, leading to better accuracy and faster processing.
- Efficient resource utilization: The dynamic allocation of hardware resources ensures that they are optimally utilized, reducing wastage and improving the overall efficiency of the system.
- Scalability and adaptability: The method allows for the operation of large neural network models and can adapt to changing requirements, making it suitable for a wide range of applications.
Original Abstract Submitted
A method for an artificial neural network operation on a plurality of subgraphs may include generating a resource determination trigger corresponding to a target subgraph among the plurality of subgraphs included in a target neural network model; generating a control signal for hardware allocated to the target subgraph and driving resource settings in response to the resource determination trigger; changing, based on the control signal, at least one of hardware allocated to the target subgraph and driving resource settings; and performing an operation on the target subgraph based on the changed hardware and driving resource settings.
