Information processing method and device

Patent metadata

Field	Value
Publication number	US-2022383064-A1
Application number	US-202117565910-A
Country	US
Kind code	A1
Filing date	Dec 30, 2021
Priority date	May 18, 2021
Publication date	Dec 1, 2022
Grant date	—

Abstract

The present disclosure discloses an information processing method and device, and relates to the field of artificial intelligence, in particular, to a graph neural network in the field of deep learning. A specific implementation solution according to an embodiment includes determining initial representation of edges connected between a plurality of atoms in a molecule based on three-dimensional structure information of the molecule; determining first representation of a neighbor edge of each of the atoms based on the initial representation of the edges, the neighbor edge of each of the atoms indicating at least one edge connected with each of the atoms; determining first representation of each of the atoms based on the first representation of the neighbor edge of each of the atoms; determining feature representation for characterizing the molecule based on the first representation of each of the atoms.

First claim

What is claimed is:

1 . An information processing method, comprising:
determining initial representation of edges connected between a plurality of atoms in a molecule based on three-dimensional structure information of the molecule; determining first representation of a neighbor edge of each of the atoms based on the initial representation of the edges, the neighbor edge of each of the atoms indicating at least one edge connected with each of the atoms; determining first representation of each of the atoms based on the first representation of the neighbor edge of each of the atoms; and determining feature representation for characterizing the molecule based on the first representation of each of the atoms.

2 . The method as claimed in claim 1 , wherein determining the first representation of the neighbor edge of each of the atoms comprises:
determining an included angle between each of the other edges except a first edge in the neighbor edge of a first atom in the plurality of atoms and the first edge to obtain a plurality of included angles; and determining the first representation of the first edge based on the plurality of included angles and initial representation of the other edges.

3 . The method as claimed in claim 2 , wherein determining the first representation of the first edge comprises:
dividing the other edges into different angle domains based on plurality of included angles; determining weighted initial representation for each of the angle domains through weighted summation of the initial representation of the other edges in each of the angle domains based on attention weight of the other edges in each of the angle domains to the first edge; and concatenating the weighted initial representation for each of the angle domains as the first representation of the first edge.

4 . The method as claimed in claim 1 , wherein determining the first representation of each of the atoms comprises:
determining a distance between the neighbor edge of the first atom in the plurality of atoms and the first atom, the distance between the neighbor edge and the first atom indicating a distance between a second atom connected with the neighbor edge and the first atom; determining the attention weight of the neighbor edge of the first atom to the first atom based on the distance; and determining an weighted average of the first representation of the neighbor edge of the first atom based on the attention weight as first representation of the first atom.

5 . The method as claimed in claim 1 , wherein determining the feature representation for characterizing the molecule comprises:
determining second representation of the edges based on the first representation of each of the atoms; determining third representation of the neighbor edge of each of the atoms based on the second representation of the edges; determining second representation of each of the atoms based on the third representation of the neighbor edge of each of the atoms; and determining the feature representation for characterizing the molecule based on the second representation of each of the atoms.

6 . The method as claimed in claim 1 , wherein the three-dimensional structure information of the molecule is represented by a polar coordinate system.

7 . The method as claimed in claim 1 , wherein determining the initial representation of edges connected between the plurality of atoms in the molecule comprises:
determining initial representation of the plurality of atoms in the molecule based on the three-dimensional structure information of the molecule; determining characterization of the distance between the plurality of atoms connected by the edge based on the three-dimensional structure information of the molecule; and determining the initial representation of edges connected between the plurality of atoms based on the initial representation of the plurality of atoms and the characterization of the distance between plurality of the atoms.

8 . The method as claimed in claim 1 , further comprising:
determining whether a distance between any two atoms of the plurality of atoms is less than a threshold distance; and in response to determining that the distance between these two atoms is less than the threshold distance, constructing an edge for connecting these two atoms.

9 . The method as claimed in claim 1 , wherein the three-dimensional structure information of the molecule at least comprises types and space distribution of the atoms forming the molecule.

10 . The method as claimed in claim 1 , wherein the initial representation of edges is one-dimensional vector, and the feature representation for characterizing the molecule is one-dimensional vector.

11 . An electronic device, comprising:
at least one processor, and a memorizer, in communication connection with the at least one processor, wherein the memorizer stores an instruction capable of being performed by the at least one processor, and the instruction is performed by the at least one processor, to cause the at least one processor to perform the following steps: determining initial representation of edges connected between a plurality of atoms in a molecule based on three-dimensional structure information of the molecule; determining first representation of a neighbor edge of each of the atoms based on the initial representation of the edges, the neighbor edge of each of the atoms indicating at least one edge connected with each of the atoms; determining first representation of each of the atoms based on the first representation of the neighbor edge of each of the atoms; and determining feature representation for characterizing the molecule based on the first representation of each of the atoms.

12 . The electronic device as claimed in claim 11 , wherein determining the first representation of the neighbor edge of each of the atoms comprises:
determining an included angle between each of the other edges except a first edge in the neighbor edge of a first atom in the plurality of atoms and the first edge to obtain a plurality of included angles; and determining the first representation of the first edge based on the plurality of included angles and initial representation of the other edges.

13 . The electronic device as claimed in claim 12 , wherein determining the first representation of the first edge comprises:
dividing the other edges into different angle domains based on plurality of included angles; determining weighted initial representation for each of the angle domains through weighted summation of the initial representation of the other edges in each of the angle domains based on attention weight of the other edges in each of the angle domains to the first edge; and concatenating the weighted initial representation for each of the angle domains as the first representation of the first edge.

14 . The electronic device as claimed in claim 11 , wherein determining the first representation of each of the atoms comprises:
determining a distance between the neighbor edge of the first atom in the plurality of atoms and the first atom, the distance between the neighbor edge and the first atom indicating a distance between a second atom connected with the neighbor edge and the first atom; determining the attention weight of the neighbor edge of the first atom to the first atom based on the distance; and determining an weighted average of the first representation of the neighbor edge of the first atom based on the attention weight as first representation of the first atom.

15 . The electronic device as claimed in claim 11 , w

Assignees

• Baidu Com Times Tech Beijing Co Ltd

Inventors

• Li Shuangli
• Zhou Jingbo
• Xu Tong
• Huang Liang
• Wang Fan
• Xiong Haoyi
• Huang Weili
• Xiong Hui
• Dou Dejing

Classifications

• G06N3/04 Primary

  Architecture, e.g. interconnection topology · CPC title

• G16C20/50 Primary

  Molecular design, e.g. of drugs · CPC title

• G16C20/70

  Machine learning, data mining or chemometrics · CPC title

• G06N3/0464

  Convolutional networks [CNN, ConvNet] · CPC title

• G06N3/08

  Learning methods · CPC title