17957508. Selecting a Tiling Scheme for Processing Instances of Input Data Through a Neural Netwok simplified abstract (ADVANCED MICRO DEVICES, INC.)
Contents
- 1 Selecting a Tiling Scheme for Processing Instances of Input Data Through a Neural Netwok
- 1.1 Organization Name
- 1.2 Inventor(s)
- 1.3 Selecting a Tiling Scheme for Processing Instances of Input Data Through a Neural Netwok - A simplified explanation of the abstract
- 1.4 Simplified Explanation
- 1.5 Potential Applications
- 1.6 Problems Solved
- 1.7 Benefits
- 1.8 Potential Commercial Applications
- 1.9 Possible Prior Art
- 1.10 Original Abstract Submitted
Selecting a Tiling Scheme for Processing Instances of Input Data Through a Neural Netwok
Organization Name
Inventor(s)
Akila Subramaniam of Allen TX (US)
Tung Chuen Kwong of Richmond Hill (CA)
Juanjo Noguera of Santa Clara CA (US)
Selecting a Tiling Scheme for Processing Instances of Input Data Through a Neural Netwok - A simplified explanation of the abstract
This abstract first appeared for US patent application 17957508 titled 'Selecting a Tiling Scheme for Processing Instances of Input Data Through a Neural Netwok
Simplified Explanation
The electronic device described in the patent application uses different tiling schemes to process input data through a neural network. Here is a simplified explanation of the abstract:
- The device selects a tiling scheme from a set of options based on information about the neural network and processing circuitry.
- Input data is divided into portions according to the selected tiling scheme.
- Each portion is processed separately in the neural network.
- The outputs from each portion are combined to generate the final output for the input data.
Potential Applications
This technology could be applied in various fields such as image recognition, natural language processing, and autonomous vehicles.
Problems Solved
This technology helps optimize the processing of input data in neural networks by using different tiling schemes, improving efficiency and accuracy.
Benefits
The use of different tiling schemes allows for more efficient processing of input data, leading to faster and more accurate results in neural networks.
Potential Commercial Applications
- "Optimizing Neural Network Processing with Tiling Schemes for Improved Efficiency and Accuracy"
Possible Prior Art
One possible prior art could be the use of parallel processing techniques in neural networks to improve efficiency and speed.
Unanswered Questions
How does this technology compare to traditional neural network processing methods?
This article does not provide a direct comparison between this technology and traditional methods of processing input data in neural networks. Further research or experimentation may be needed to determine the advantages and disadvantages of using tiling schemes.
What are the specific properties of the tiling schemes used in this technology?
The abstract does not delve into the specific properties of the tiling schemes selected by the electronic device. Understanding the characteristics of these tiling schemes could provide insights into their effectiveness and potential limitations.
Original Abstract Submitted
An electronic device uses a tiling scheme selected from among a set of tiling schemes for processing instances of input data through a neural network. Each of the tiling schemes is associated with a different arrangement of portions into which instances of input data are divided for processing in the neural network. In operation, processing circuitry in the electronic device acquires information about a neural network and properties of the processing circuitry. The processing circuitry then selects a given tiling scheme from among a set of tiling schemes based on the information. The processing circuitry next processes instances of input data in the neural network using the given tiling scheme. Processing each instance of input data in the neural network includes dividing the instance of input data into portions based on the given tiling scheme, separately processing each of the portions in the neural network, and combining the respective outputs to generate an output for the instance of input data.
