View source for 18112769. NEURAL NETWORK COMPUTING SYSTEM AND METHOD OF EXECUTING NEURAL NETWORK MODEL simplified abstract (SAMSUNG ELECTRONICS CO., LTD.)

=NEURAL NETWORK COMPUTING SYSTEM AND METHOD OF EXECUTING NEURAL NETWORK MODEL=


==Organization Name==

[[:Category:SAMSUNG ELECTRONICS CO., LTD.|SAMSUNG ELECTRONICS CO., LTD.]]


[[Category:SAMSUNG ELECTRONICS CO., LTD.]]

==Inventor(s)==

[[:Category:Jungho Kim of Suwon-si (KR)|Jungho Kim of Suwon-si (KR)]][[Category:Jungho Kim of Suwon-si (KR)]]

[[:Category:Hyunjin Kim of Suwon-si (KR)|Hyunjin Kim of Suwon-si (KR)]][[Category:Hyunjin Kim of Suwon-si (KR)]]

[[:Category:Youngchan Cho of Suwon-si (KR)|Youngchan Cho of Suwon-si (KR)]][[Category:Youngchan Cho of Suwon-si (KR)]]

[[:Category:Hoon Choi of Suwon-si (KR)|Hoon Choi of Suwon-si (KR)]][[Category:Hoon Choi of Suwon-si (KR)]]

==NEURAL NETWORK COMPUTING SYSTEM AND METHOD OF EXECUTING NEURAL NETWORK MODEL - A simplified explanation of the abstract==


This abstract first appeared for US patent application 18112769 titled 'NEURAL NETWORK COMPUTING SYSTEM AND METHOD OF EXECUTING NEURAL NETWORK MODEL

==Simplified Explanation==

The abstract describes a neural network computing system that includes a processor with heterogeneous computing devices, a memory for buffering input and output data, a memory controller for controlling data input and output, and a system bus for communication between the processor and memory controller. The processor determines target execution times for each node in the neural network model and controls operating frequencies of hardware devices based on the target computing device, amount of work, and target execution time for each node.

* The system includes a processor with heterogeneous computing devices.
* A memory is used to buffer input and output data of the neural network model.
* A memory controller controls data input and output of the memory.
* A system bus supports communication between the processor and memory controller.
* The processor determines target execution times for each node in the neural network model.
* Operating frequencies of hardware devices are controlled based on the target computing device, amount of work, and target execution time for each node.
* The neural network model is executed by operating at the determined operating frequencies.

Potential applications of this technology:
* Artificial intelligence and machine learning systems
* Image and speech recognition
* Natural language processing
* Autonomous vehicles
* Robotics

Problems solved by this technology:
* Optimization of neural network execution time
* Efficient utilization of hardware devices
* Improved performance and accuracy of neural network models

Benefits of this technology:
* Faster execution of neural network models
* Enhanced efficiency and resource utilization
* Improved accuracy and performance of AI systems


==Original Abstract Submitted==

A neural network computing system includes a processor including heterogeneous computing devices configured to execute a neural network model; a memory configured to buffer input data and output data of the neural network model; a memory controller configured to control data input and data output of the memory; and a system bus configured to support communication between the processor and the memory controller. The processor determines a target execution time for each node included in the neural network model based on a target end-to-end execution time of the neural network model; controls operating frequencies of hardware devices including the heterogeneous computing devices, the memory controller, and the system bus based on a target computing device for execution of each node, an amount of work for each node, and the target execution time for each node, and executes the neural network model by operating at the operating frequencies.



[[Category:G06F11/34]]
[[Category:G06N3/10]]