18148304. DATA FEATURE EXTRACTION METHOD AND RELATED APPARATUS simplified abstract (Huawei Technologies Co., Ltd.)
Contents
DATA FEATURE EXTRACTION METHOD AND RELATED APPARATUS
Organization Name
Inventor(s)
DATA FEATURE EXTRACTION METHOD AND RELATED APPARATUS - A simplified explanation of the abstract
This abstract first appeared for US patent application 18148304 titled 'DATA FEATURE EXTRACTION METHOD AND RELATED APPARATUS
Simplified Explanation
The patent application describes a method and apparatus for extracting data features in the field of artificial intelligence. The method involves performing an additional convolution operation on quantized data to extract a target feature. This is done by calculating the sum of absolute differences between the quantized feature extraction parameters and the quantized data. The target feature is then obtained based on this sum. The feature extraction parameters and data are quantized using the same quantization parameter.
- The method involves performing an additional convolution operation on quantized data to extract a target feature.
- The target feature is obtained by calculating the sum of absolute differences between the quantized feature extraction parameters and the quantized data.
- The feature extraction parameters and data are quantized using the same quantization parameter.
Potential applications of this technology:
- Artificial intelligence applications on resource-limited devices.
- Data feature extraction in various fields such as image recognition, natural language processing, and speech recognition.
Problems solved by this technology:
- Saves storage and computing resources by reducing the limitations of resource-limited devices in applying artificial intelligence.
- Provides a method for efficient feature extraction from quantized data.
Benefits of this technology:
- Saves storage and computing resources.
- Enables the application of artificial intelligence on resource-limited devices.
- Provides an efficient method for feature extraction from quantized data.
Original Abstract Submitted
A data feature extraction method and apparatus in the field of artificial intelligence are provided. An addition convolution operation is performed to extract a target feature in quantized data based on quantized feature extraction parameters, that is, to calculate a sum of absolute values of differences between the quantized feature extraction parameters and the quantized data, to obtain the target feature based on the sum. In addition, feature extraction parameters and data are quantized by using a same quantization parameter. According to this application, a storage resource is saved, a computing resource is saved, thereby reducing a limitation on an application of artificial intelligence to a resource-limited device. Further, when the extracted feature data is dequantized, the feature data may be dequantized based on the quantization parameters.
