Online model optimization method and device, equipment, storage medium and program product
An optimization method and model technology, applied in storage media and program products, online model optimization methods, equipment, devices, and artificial intelligence fields, can solve the problems of complex sensitivity of detection model features, complex and inefficient model optimization process, etc. Achieve the effect of reducing complexity and improving efficiency
Pending Publication Date: 2021-06-29
BEIJING BAIDU NETCOM SCI & TECH CO LTD
View PDF0 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0003] At present, when optimizing machine learning models in the industry, the sensitivity of the model to features is detected from the feature dimension level, but for complex combined feature models, cross-feature models, etc., the sensitivity of the detection model to features is more complicated, resulting in the model The optimization process is complex and inefficient
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0030] The exemplary embodiments of the present application will be described below in connection with the accompanying drawings, including various details of the present application embodiments to help understand, and they should be considered simply exemplary. Accordingly, it will be appreciated by those skilled in the art that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the present application. Similarly, a description of the well-known functions and structures is omitted in order to clear and concise, the following description is omitted.
[0031] Machine learning is a multi-field cross-discipline. The core problem of its concern is to obtain rules from historical experience, and use it into new similar scenarios. He hereby historical experience and how to obtain regularity and use involves data. (Sample) and the selection and use of data. For machine learning models, not only require models t...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention discloses an online model optimization method and device, equipment, a storage medium and a program product, and relates to the field of artificial intelligence, machine learning and content recommendation. The method comprises the following steps: receiving a first message, wherein the first message comprises identification information of an online to-be-optimized model; obtaining original sample data, and transforming fields in the original sample data to obtain transformed sample data; according to the identification information of the online to-be-optimized model, constructing a construction model which is the same as the online to-be-optimized model; and optimizing the online to-be-optimized model according to the original sample data, the transformed sample data and the construction model. According to the method, the model optimization complexity is reduced, and the model optimization efficiency is improved.
Description
Technical field [0001] The present application embodiments relate to artificial intelligence technologies, in particular to an optimization method, apparatus, device, storage medium, and program product of online model, can be used in machine learning, content recommendation. Background technique [0002] For machine learning models, not only require models to achieve good predictive capabilities for existing data, but more importantly, it is possible to have good predictive capabilities for other unknown data, to improve the prediction of machine learning models. Accuracy, you need to continue to advance the optimization iteration of the model, and provide more precise services through continuous adjustment. [0003] At present, when the machine learning model is optimized, it is the sensitivity of the model to characteristics from the feature dimensional level, but for complex combination characteristics, cross-feature models, etc., the detection model is more complicated, resu...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More IPC IPC(8): G06N20/00
CPCG06N20/00
Inventor 杨婉璐李小聪陈葳蕤魏龙王召玺王峰
Owner BEIJING BAIDU NETCOM SCI & TECH CO LTD