18349124. NEURAL NETWORK PROCESSING simplified abstract (Arm Limited)
Contents
NEURAL NETWORK PROCESSING
Organization Name
Inventor(s)
Thomas James Cooksey of Cambridgeshire (GB)
Elliot Maurice Simon Rosemarine of London (GB)
NEURAL NETWORK PROCESSING - A simplified explanation of the abstract
This abstract first appeared for US patent application 18349124 titled 'NEURAL NETWORK PROCESSING
The abstract describes a data processing system with a processor capable of neural network processing, execution units for processing operations, a control circuit for task distribution, and a graphics processor with a programmable execution unit for processing programs.
- The processor is configured to perform neural network processing with execution units for processing operations.
- The control circuit distributes processing tasks to the execution unit or units.
- The graphics processor includes a programmable execution unit that can execute processing programs.
- In response to specific neural network processing tasks, the control circuit causes the programmable execution unit of the graphics processor to execute the necessary program.
Potential Applications: - Artificial intelligence - Machine learning - Image and speech recognition - Autonomous vehicles - Robotics
Problems Solved: - Efficient neural network processing - Task distribution optimization - Enhanced graphics processing capabilities
Benefits: - Faster processing of neural networks - Improved performance in AI applications - Enhanced graphics rendering - Increased efficiency in processing tasks - Potential for new innovations in technology
Commercial Applications: The technology could be utilized in various industries such as healthcare, finance, automotive, and entertainment for applications like medical diagnosis, fraud detection, autonomous driving, and gaming.
Questions about the technology: 1. How does the integration of a graphics processor enhance neural network processing? 2. What are the potential limitations of using a programmable execution unit for neural network processing?
Original Abstract Submitted
A data processing system comprising a processor () that is configured to perform neural network processing having one or more execution units () configured to perform processing operations for neural network processing and a control circuit () configured to distribute processing tasks to the execution unit or units, and a graphics processor () comprising a programmable execution unit () operable to execute processing programs to perform processing operations. The control circuit () of the processor () that is configured to perform neural network processing is configured to, in response to an indication of particular neural network processing to be performed provided to the control circuit, cause the programmable execution unit () of the graphics processor to execute a program to perform the indicated neural network processing.
