View source for Qualcomm incorporated (20240338555). METHOD AND APPARATUS FOR UTILIZING EXTERNAL NEURAL PROCESSOR FROM GRAPHICS PROCESSOR simplified abstract

=METHOD AND APPARATUS FOR UTILIZING EXTERNAL NEURAL PROCESSOR FROM GRAPHICS PROCESSOR=


==Organization Name==

[[:Category:qualcomm incorporated|qualcomm incorporated]]


[[Category:qualcomm incorporated]]

==Inventor(s)==

[[:Category:Elina Kamenetskaya of Belmont MA (US)|Elina Kamenetskaya of Belmont MA (US)]][[Category:Elina Kamenetskaya of Belmont MA (US)]]

[[:Category:Amir Momeni of Arlington MA (US)|Amir Momeni of Arlington MA (US)]][[Category:Amir Momeni of Arlington MA (US)]]

[[:Category:Hamza Omar of San Diego CA (US)|Hamza Omar of San Diego CA (US)]][[Category:Hamza Omar of San Diego CA (US)]]

[[:Category:Engin Ipek of San Diego CA (US)|Engin Ipek of San Diego CA (US)]][[Category:Engin Ipek of San Diego CA (US)]]

[[:Category:Alexei Vladimirovich Bourd of San Diego CA (US)|Alexei Vladimirovich Bourd of San Diego CA (US)]][[Category:Alexei Vladimirovich Bourd of San Diego CA (US)]]

[[:Category:Zifeng Li of San Diego CA (US)|Zifeng Li of San Diego CA (US)]][[Category:Zifeng Li of San Diego CA (US)]]

==METHOD AND APPARATUS FOR UTILIZING EXTERNAL NEURAL PROCESSOR FROM GRAPHICS PROCESSOR - A simplified explanation of the abstract==


This abstract first appeared for US patent application 20240338555 titled 'METHOD AND APPARATUS FOR UTILIZING EXTERNAL NEURAL PROCESSOR FROM GRAPHICS PROCESSOR

'''Simplified Explanation:'''
The patent application describes a system for concurrent tensor processing using multiple processing engines, where one engine accesses input and kernel tensors from a common memory unit, and another engine generates an output tensor based on these tensors.

'''Key Features and Innovation:'''
* Apparatus with a common memory unit
* First processing engine accessing input and kernel tensors
* Second processing engine generating output tensor based on input and kernel tensors

'''Potential Applications:'''
This technology could be used in various fields such as artificial intelligence, machine learning, image processing, and data analysis.

'''Problems Solved:'''
This technology addresses the need for efficient concurrent tensor processing, improving overall processing speed and performance.

'''Benefits:'''
* Faster processing of tensors
* Improved efficiency in handling complex data
* Enhanced performance in various applications

'''Commercial Applications:'''
Potential commercial uses include implementing this technology in AI systems, computer vision applications, and big data analytics, leading to faster and more accurate results.

'''Prior Art:'''
Researchers interested in this technology may want to explore prior art related to concurrent tensor processing, multiple processing engines, and memory unit configurations in the field of computer science and artificial intelligence.

'''Frequently Updated Research:'''
Stay updated on the latest advancements in concurrent tensor processing, multiple processing engines, and memory unit technologies to enhance the efficiency and performance of tensor processing systems.

'''Questions about concurrent tensor processing with multiple processing engines:'''
1. How does this technology improve the efficiency of tensor processing?
2. What are the potential applications of concurrent tensor processing in various industries?

By providing a detailed and comprehensive overview of the patent application, readers can gain a better understanding of the innovative technology and its potential impact on various industries.


==Original Abstract Submitted==

aspects of the disclosure are directed to concurrent tensor processing with multiple processing engines. in accordance with one aspect, an apparatus including a common memory unit; a first processing engine coupled to the common memory unit, wherein the first processing engine is configured to access a portion of an input tensor and a portion of a kernel tensor from the common memory unit; and a second processing engine coupled to the common memory unit, wherein the first processing engine is further configured to send the portion of the input tensor and the portion of the kernel tensor to the second processing engine and wherein the second processing engine is configured to generate a portion of an output tensor based on the portion of the input tensor and on the portion of the kernel tensor.



[[Category:G06N3/063]]
[[Category:CPC_G06N3/063]]