Search

EP-4742254-A1 - SYSTEM, DEVICE AND METHOD FOR MATERIAL SUBGRAPH MODEL

EP4742254A1EP 4742254 A1EP4742254 A1EP 4742254A1EP-4742254-A1

Abstract

A system and a method are disclosed for material property prediction using subgraph modeling. A method may utilize artificial intelligence (Al)-based material subgraph models, including the decomposition of the molecular structure of materials, to generate graphs of subgraphs, and implement subgraph modeling to scale AI-based models for use with large and/or complex molecules and enhance the material property predictions. Aspects can involve utilizing the enhanced material property predictions to improve the overall performance of display related products. For instance, by using material property predictions with greater accuracy from subgraph modeling, materials deemed most suitable and/or efficient may be used to manufacture a display device. The method may include generating subgraphs comprising a molecular substructure of a material, and applying an AI-based model to the subgraphs to generate a material property prediction based on the molecular substructure of the material.

Inventors

  • QU, Shuhui
  • PARK, CHEOL WOO
  • CHENG, Qisen
  • LEE, JANGHWAN

Assignees

  • Samsung Display Co., Ltd.

Dates

Publication Date
20260513
Application Date
20250930

Claims (15)

  1. A method comprising: generating, by a processor (212), subgraphs (202), each of the subgraphs (202) comprising a molecular substructure of a material (201); applying, by the processor, an artificial intelligence (AI)-based model to the subgraphs to generate a material property prediction (236) based on the molecular substructure of the material (201); determining, by the processor (212), a function related to the material for production of a target device based on the material property prediction; and transmitting, by the processor (212), a signal to a component (102) to control the component to execute the function related to the material (201) for the production of the target device.
  2. The method of claim 1, further comprising decomposing a graph of a molecular structure of the material into the molecular substructures.
  3. The method of claim 2, wherein the decomposing comprises Breaking Retrosynthetically Interesting Chemical bonds (BRIC) decomposition.
  4. The method of any preceding claim, further comprising generating embeddings (210) of the subgraphs (202) based on the subgraphs (202).
  5. The method of claim 4, wherein the generating the embeddings (210) of the subgraphs (202) comprises processing the subgraphs (202) by a graph neural network (215).
  6. The method of claim 4 or claim 5, further comprising generating a graph of subgraphs (220) based on the embeddings (210) of the subgraphs (202).
  7. The method of claim 6, wherein the graph of subgraphs (220) comprises nodes (221) and edges (222).
  8. The method of claim 7, wherein each of the nodes (221) correspond to one of the subgraphs (202), and a value for each of the nodes (221) correspond to one of the embeddings (210) of the subgraphs (202).
  9. The method of claim 8, wherein each of the edges (222) represent a relationship between the subgraphs (202).
  10. The method of any one of claims 6 - 9, further comprising generating updated node embeddings (605) based on the graph of subgraphs; wherein, optionally, the generating the updated node embeddings (605) comprises processing the graph of subgraphs (220) by a graph neural network (230).
  11. The method of claim 10, wherein the AI-based model analyzes the graph of subgraphs (220) and models the relationships between the subgraphs (220).
  12. The method of any preceding claim, wherein the target device comprises an organic light-emitting diode (OLED) display device.
  13. The method of any preceding claim, wherein the material property prediction (236) is based one or more of: a physical property of the material, a chemical property of the material, mechanical property of the material, or an optical property of the material.
  14. A device comprising: one or more processors (212) that are configured to perform a method according to any preceding claim.
  15. A system (102) comprising a device according to claim 14, and a memory (211) storing instructions, which, based on being executed by the processing circuit, cause the processing circuit to perform the method according to any one of claims 1 - 13.

Description

BACKGROUND 1. FIELD Aspects of some embodiments of the present disclosure generally relate to machine learning and/or artificial intelligence. More particularly, the subject matter disclosed herein relates to determining material properties for display related products based on artificial intelligence. 2. DESCRIPTION OF THE RELATED ART Material property prediction may involve estimating the physical, chemical, mechanical, and/or optical properties of materials that can be used in display related products. Therefore, it may be desirable to utilize computational models, for example artificial intelligence-based models, to generate material property predictions that can then be utilized in various aspects of the design and/or manufacture of display related products, including material selection, performance optimization, and innovation in display technology. The above information disclosed in this Background section is for enhancement of understanding of the background of the present disclosure, and therefore, it may contain information that does not constitute prior art. TECHNICAL FIELD The disclosure generally relates to electronic devices. More particularly, the subject matter disclosed herein relates to determining material properties for display related electronic devices based on artificial intelligence. SUMMARY Aspects of some embodiments of the present disclosure relate to material property prediction and/or analysis. For example, aspects of some embodiments of the present disclosure relate to improvements to the accuracy and efficiency of material property predictions by utilizing artificial intelligence and/or material subgraph models. Material property prediction may involve estimating the physical, chemical, mechanical, and/or optical properties of materials that can be used in display related products. Therefore, it may be desirable to utilize computational models, for example artificial intelligence-based models, to generate material property predictions that can then be utilized in various aspects of the design and/or manufacture of display related products, including material selection, performance optimization, and innovation in display technology. However, there may be issues associated with applying artificial intelligence (AI) techniques to material property prediction, in a manner that maintains the robustness and/or efficiency of AI and can scale to the relatively large size and complexity of the molecular structure of materials that may be used for display related products. Aspects of some embodiments of the present disclosure relate to systems and methods for AI-based material subgraph models, including the decomposition of the molecular structure of materials, generating graphs of subgraphs, and implementing subgraph modeling in a manner that may scale AI-based models to large and/or complex molecules and enhances their expressive power. Thus, the disclosed embodiments may improve the overall performance of display related products by utilizing materials deemed most suitable and/or efficient. In some embodiments, a method includes: generating, by a processor, subgraphs, each of the subgraphs including a molecular substructure of a material; applying, by the processor, an AI-based model to the subgraphs to generate a material property prediction based on the molecular substructure of the material; determining, by the processor, a function related to the material for production of a target device based on the material property prediction; and transmitting, by the processor, a signal to a component to control the component to execute the function related to the material for the production of the target device. In some embodiments, the method may further include decomposing a graph of a molecular structure of the material into the molecular substructures. In some embodiments, the decomposing may include Breaking Retrosynthetically Interesting Chemical bonds (BRIC) decomposition. In some embodiments, the method may further include generating embeddings of the subgraphs based on the subgraphs. In some embodiments, generating the embeddings of the subgraphs may include processing the subgraphs by a graph neural network. In some embodiments, the method may further include generating a graph of subgraphs based on the embeddings of the subgraphs. In some embodiments, the graph of subgraphs may include nodes and edges. In some embodiments, each of the nodes correspond to one of the subgraphs and a value for each of the nodes correspond to one of the embeddings of the subgraphs. In some embodiments, each of the edges represent a relationship between the subgraphs. In some embodiments, the method may further include generating updated node embeddings based on the graph of subgraphs. In some embodiments, generating the updated node embeddings may include processing the graph of subgraphs by a graph neural network. In some embodiments, the AI-based model analyzes the graph of subgraphs and models the