Learning device, inference device, learning model generation method and inference method
A technology for learning models and learning devices, applied in neural learning methods, computing models, machine learning, etc.
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
no. 1 Embodiment approach
[0032]
[0033] First, the overall configuration of a simulation system for simulating a semiconductor manufacturing process will be described. figure 1 It is a schematic diagram of an example of the overall structure of the simulation system. Such as figure 1 As shown, the simulation system 100 has a learning device 110 , an inference device 120 and a simulation device 130 .
[0034] A data shaping program and a learning program are installed in the learning device 110 , and by executing the program, the learning device 110 can function as a data shaping unit 111 and a learning unit 112 .
[0035] The data shaping unit 111 is an example of a processing unit. The data shaping unit 111 reads the learning data transmitted from the simulation device 130 and stored in the learning data storage unit 113, processes part of the read learning data into a shape suitable for the learning unit 112, and inputs it into the learning model. the scheduled form.
[0036] The learning uni...
no. 2 Embodiment approach
[0169] In the first embodiment described above, the parameter data is processed into image data according to the vertical size and horizontal size of the image data before processing, and then connected with the image data before processing, and then input into the learning model (or learned model) .
[0170] However, the method of processing parameter data and the method of inputting processed parameter data into a learning model (or a learned model) are not limited thereto. For example, the processed parameter data may also be configured as input to each layer of the learning model (or the learned model). In addition, the parameter data can be configured so that when each layer of the learning model (or the learned model) is input, it is processed into an image that is convoluted by each layer of the learning model (or the learned model). A predetermined form in which data is transformed. Next, the second embodiment will be described focusing on the points of difference fr...
no. 3 Embodiment approach
[0196] In the above-described first and second embodiments, when the learning unit performs machine learning, no particular mention is made of phenomena unique to semiconductor manufacturing processes. On the other hand, there are unique phenomena in the semiconductor manufacturing process, and by reflecting them to the machine learning based on the learning part (that is, by reflecting domain knowledge to the machine learning based on the learning part), it is possible to further improve Analog Accuracy. Next, a third embodiment reflecting domain knowledge will be described focusing on differences from the first and second embodiments described above.
[0197]
[0198] Figure 14 It is a schematic diagram of an example of the functional configuration of the learning unit of the learning device according to the third embodiment. The internal composition of the learning model and the Figure 4 The illustrated learning unit 112 has a different functional configuration. It ...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More - R&D
- Intellectual Property
- Life Sciences
- Materials
- Tech Scout
- Unparalleled Data Quality
- Higher Quality Content
- 60% Fewer Hallucinations
Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.
© 2025 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com