WO2021096009A1 - Procédé et dispositif permettant d'enrichir la connaissance sur la base d'un réseau de relations - Google Patents

Procédé et dispositif permettant d'enrichir la connaissance sur la base d'un réseau de relations Download PDF

Info

Publication number
WO2021096009A1
WO2021096009A1 PCT/KR2020/006239 KR2020006239W WO2021096009A1 WO 2021096009 A1 WO2021096009 A1 WO 2021096009A1 KR 2020006239 W KR2020006239 W KR 2020006239W WO 2021096009 A1 WO2021096009 A1 WO 2021096009A1
Authority
WO
WIPO (PCT)
Prior art keywords
relation network
knowledge
node
path
relationship
Prior art date
Application number
PCT/KR2020/006239
Other languages
English (en)
Korean (ko)
Inventor
박영택
이완곤
바트셀렘자그바랄
노재승
Original Assignee
숭실대학교산학협력단
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 숭실대학교산학협력단 filed Critical 숭실대학교산학협력단
Publication of WO2021096009A1 publication Critical patent/WO2021096009A1/fr

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N5/00—Computing arrangements using knowledge-based models
    • G06N5/02—Knowledge representation; Symbolic representation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/901—Indexing; Data structures therefor; Storage structures
    • G06F16/9024—Graphs; Linked lists

Definitions

  • the present invention relates to a method and apparatus for supplementing knowledge on a knowledge graph using a relation network (RN).
  • RN relation network
  • Knowledge Graphs are used as important resources in the fields of machine learning and data mining, and are particularly useful for solving problems such as question answering, fact checking, and link prediction.
  • the knowledge graph is a knowledge network composed of entity nodes and relationship edges, and may be expressed as a triple ⁇ h, r, t> in an RDF format. In this case, h is the head entity, and r is the relationship between the tail entity t connected to h.
  • knowledge graphs are widely used in various tasks, there is a problem that correctness and completeness are not guaranteed.
  • KGC Knowledge Graph Completion
  • Korean Patent Laid-Open Publication No. 10-2016-0064826 title of the invention: an apparatus and method for providing a semantic search service based on a knowledge graph, publication date: June 8, 2016).
  • An embodiment of the present invention is to provide a method and apparatus for supplementing knowledge showing excellent performance while solving the problem of the existing KGC by extracting a relationship path based on a Path Ranking Algorithm (PRA) and using it as training data for a relation network. .
  • PRA Path Ranking Algorithm
  • a knowledge supplementation method based on a relation network for achieving the above object is a plurality of nodes representing a relationship between a source node constituting a node pair and a target node for a plurality of node pairs included in the knowledge graph. Extracting route information, which is information about the route of Generating training data corresponding to each of the plurality of paths based on the path information; And training a relation network model using the training data.
  • the training data includes a context including information on a relationship represented by each of at least one link constituting an individual path, a question about a relationship between a source node and a target node, and the question. May include an answer to.
  • a source node, a target node, and a first triple composed of a relationship between the source node and the target node of each of the at least one link are converted into a long A Short (LSTM).
  • LSTM long A Short
  • the at least one link may be less than or equal to a predetermined threshold number.
  • the step of extracting the path information may extract the plurality of paths based on a Path Ranking Algorithm (PRA).
  • PRA Path Ranking Algorithm
  • the knowledge supplement apparatus based on a relation network for achieving the above-described object provides a relationship between a source node and a target node constituting a node pair with respect to a plurality of node pairs included in the knowledge graph.
  • a route extraction unit for extracting route information, which is information on a plurality of routes shown;
  • a data generator for generating training data corresponding to each of the plurality of paths based on the path information;
  • a learning unit that trains a relation network model by using the training data.
  • the training data includes a context including information on a relationship represented by each of at least one link constituting an individual path, a question about a relationship between a source node and a target node, and the question. May include an answer to.
  • the learning unit encodes the source node of each of the at least one link, the target node, and the relationship between the source node and the target node into Long A Short-Term Memory (LSTM), and encodes the encoded result.
  • LSTM Long A Short-Term Memory
  • the learning unit encodes the source node of each of the at least one link, the target node, and the relationship between the source node and the target node into Long A Short-Term Memory (LSTM), and encodes the encoded result.
  • LSTM Long A Short-Term Memory
  • the at least one link may be less than or equal to a predetermined threshold number.
  • the path extraction unit may extract the plurality of paths based on a Path Ranking Algorithm (PRA).
  • PRA Path Ranking Algorithm
  • the method and apparatus for supplementing knowledge based on a relation network extracts a relational path based on a PRA (Path Ranking Algorithm) and uses it as training data for the relation network, thereby solving the problem of the existing KGC and showing excellent performance. have.
  • PRA Pulth Ranking Algorithm
  • the knowledge supplement method and apparatus based on the relation network facilitates extraction of meaningful information such as customized services specialized for a user, and thus various service fields of artificial intelligence (Q&A system, recommendation system, interactive agent system, etc. ), there is an effect that can be used.
  • Q&A system Q&A system, recommendation system, interactive agent system, etc.
  • FIG. 1 is a flowchart illustrating a method of supplementing knowledge based on a relation network according to an embodiment of the present invention.
  • FIG. 2 is a flowchart illustrating a method of learning a relation network model according to an embodiment of the present invention.
  • FIG. 3 is a block diagram illustrating an apparatus for supplementing knowledge based on a relation network according to an embodiment of the present invention.
  • FIG. 4 is a diagram showing a path matrix according to an embodiment of the present invention.
  • FIG. 5 is a diagram illustrating a path sequence according to an embodiment of the present invention.
  • FIG. 6 is a diagram for explaining training data according to an embodiment of the present invention.
  • FIG. 7 is a diagram illustrating a learning process according to an embodiment of the present invention.
  • FIG. 1 is a flowchart illustrating a method of supplementing knowledge based on a relation network according to an embodiment of the present invention.
  • step S110 the knowledge supplement apparatus provides path information, which is information on a plurality of paths indicating a relationship between a source node and a target node constituting the node pair for a plurality of node pairs included in the knowledge graph. Extract.
  • the knowledge supplement device extracts path information, which is information about a plurality of paths representing the relationship between the source node and the target node, for a plurality of node pairs that pair two nodes among them. can do.
  • the knowledge supplement apparatus may include path information including information on a plurality of paths existing between A and B with respect to the source node A and the target node B. More specifically, when A and B are connected through intermediate nodes C and D respectively, the knowledge supplement apparatus may extract information about the paths of A-C-B and A-D-B as path information.
  • the relationship between the source node and the target node may include content describing the relationship between the subject and the object when the source node is a subject and the target node is an object. For example, when the source node is lebron james and the target node is LA lakers, the relationship may be playsFor.
  • the knowledge supplement device may extract a plurality of paths based on a Path Ranking Algorithm (PRA).
  • PRA Path Ranking Algorithm
  • the knowledge supplement apparatus may extract a plurality of paths for the source node and the target node from the knowledge graph using the PRA.
  • the knowledge supplement device can use a random walk on graph algorithm, and the random walk on graph algorithm starts from the source node and moves through other nodes in the middle to reach the target node. Algorithm.
  • the knowledge supplement apparatus may generate a path matrix by calculating a random walk probability for all paths for a plurality of node pairs.
  • each cell value of the path matrix refers to the probability that the source node s i reaches the target node t i through the path ⁇ i (the i-th column). It becomes possible to identify routes that are not helpful. For example, ⁇ 2 of FIG. 4 may be classified as a poor path because it has a relatively low probability for most of the node pairs (s i , t i ) compared to ⁇ 1. Alternatively, even if the probability for some node pairs is high, such as ⁇ 1 , it can be determined that it is difficult to classify a path that is helpful for learning even if the majority of node pairs are not connected or have a low probability.
  • the knowledge supplement device may select paths in which the ratio of node pairs connected through each path occupies more than 70% of the total. For example, if the proportion connected through path ⁇ 1 among all triples for a given relationship R is less than 50%, the column for path ⁇ 1 can be excluded from the path matrix.
  • the knowledge supplement apparatus may select paths to which most of the nodes existing in the node pair can be connected except for paths with a cell value of less than 5% on average in the path matrix, which is a random walk probability for each path.
  • step S120 the knowledge supplement apparatus generates training data corresponding to each of the plurality of routes based on the route information.
  • the knowledge supplement apparatus may generate training data for each of a plurality of paths corresponding to an individual node pair.
  • the training data includes a context including information on a relationship represented by each of at least one link constituting an individual path, a question about a relationship between a source node and a target node, and the question. May include an answer to.
  • RN relation network
  • the context may include information on at least one link corresponding to a relationship sequence existing between the source node and the target node.
  • the relationship sequence may be divided into individual relationship units (ie, link units) constituting the relationship sequence, and may be divided into a source node, a target node, and a triple structure comprising the relationship. That is, the context can be composed of the separate triple set.
  • the question can be generated using a target relation with the source node of the first triple of the triple set.
  • the relationship sequence extracted through the PRA becomes playsFor ⁇ worksFor -1 ⁇ playsIn.
  • lebron james and NBA can be matched to Athlete and League, and through this, the knowledge supplement device can generate questions and answers included in the training data as ⁇ leb-ron_james playsIn ?> and NBA.
  • the knowledge supplement apparatus may extract a context for a question in the form of a triple from an instance matching the relationship sequence.
  • ⁇ lebron james playsFor LA lakers>
  • ⁇ LA lakers playsFor -1 rajon rondo>
  • ⁇ rajon rondo playsIn NBA>
  • a story composed of a context, a question, and an answer from the training data may be constructed as shown in FIG. 6, and may be used as training data for learning of the RN.
  • At least one link may be less than or equal to a predetermined threshold number.
  • the knowledge supplementation apparatus may ensure that each of the plurality of paths includes only links less than a threshold number. This is because, when the number of links included in the path exceeds the critical number, the amount of computation required for the knowledge supplement device may increase in proportion thereto.
  • the knowledge supplement apparatus may extract only paths including only three or fewer links.
  • step S130 the knowledge supplement device learns the relation network model by using the training data.
  • the relation network is proposed by DeepMind and is a deep learning-based learning model that infers the relationship between objects.
  • RN is composed of a structure that uses training data in the form of a story consisting of context, question, and answer as input, and learns the model through two multi-layer perceptrons (MLPs).
  • MLPs multi-layer perceptrons
  • Equation 1 the relation network model can be expressed using Equation 1 below.
  • o i and o j are the i and j-th objects, respectively, a is the answer, q is the question, Is the relation function, Is a parameter that predicts an answer to a question based on the learned relationship information.
  • the combination pair (o i , o j ) of individual sentences (ie, source node, target node, and their relationship) constituting the context is merged with the question q, and the first MLP is Relationships can be learned through. Also, the second MLP It is possible to learn a parameter that predicts an answer to a question based on the relationship information learned through.
  • the knowledge supplement method based on the relation network extracts a relational path based on a PRA (Path Ranking Algorithm) and uses it as training data of the relation network, thereby solving the problem of the existing KGC It has the effect of indicating performance.
  • PRA Pulth Ranking Algorithm
  • FIG. 2 is a flowchart illustrating a method of learning a relation network model according to an embodiment of the present invention.
  • step S210 the knowledge supplement device inputs a source node, a target node, and a first triple consisting of a relationship between the source node and the target node of each of the at least one link into a Long A Short-Term Memory (LSTM). Encode.
  • LSTM Long A Short-Term Memory
  • the path may include three first triples.
  • the three first triples are (h, R 1 , e 1 ), (e 1 , R 2 , e 2 ), (e 2 , R 3 , e 3 ).
  • the knowledge supplement apparatus may obtain C 1 , C 2 , and C 3 respectively as a result of encoding the three first triples into the LSTM.
  • step S220 the knowledge supplement device generates a first result vector by first learning a relation network model using two of the encoded results and a plurality of second triples consisting of questions.
  • the knowledge supplement device selects two of the encoded results C 1 , C 2 , and C 3 , and inputs the question q into an LSTM to generate a total of three second triples including the result of encoding.
  • the first result vector can be generated by learning by using it as an input of.
  • step S230 the knowledge supplement apparatus adds the first result vector in element units to secondarily learn the relation network model.
  • the knowledge supplement device sums the first result vector in an element-wise sum and is included in the relation network model. It can be learned by using it as an input of.
  • each layer can consist of 256 units. All input data It can be considered that the context and the question are embedded together as it passes through. After each The first result vectors of are Is used as the input of. There are a total of 3 fully connected layers, and the first layer may consist of 256 units and the second layer may consist of 512 units. The last layer is set to the overall vocabulary size, so the softmax value for the answer can be output.
  • the knowledge supplement apparatus may predict a relationship between missing nodes by using the learned relation network model. Furthermore, the knowledge supplement device can provide various services based on artificial intelligence by applying the learned relation network model to a Q&A system, a recommendation system, an interactive agent system, a chatbot system, and the like.
  • FIG. 3 is a block diagram illustrating an apparatus for supplementing knowledge based on a relation network according to an embodiment of the present invention.
  • a knowledge supplement device 300 based on a relation network includes a path extraction unit 310, a data generation unit 320, and a learning unit 330.
  • the knowledge supplement device 300 based on a relation network may be mounted on a desktop PC, a notebook PC, a smart phone, a tablet PC, and a server computer.
  • the path extracting unit 310 extracts path information, which is information about a plurality of paths representing a relationship between a source node and a target node constituting the node pair, for a plurality of node pairs included in the knowledge graph.
  • the data generator 320 generates training data corresponding to each of the plurality of paths based on the path information.
  • the learning unit 330 trains the relation network model by using the training data.
  • the training data includes a context including information on a relationship represented by each of at least one link constituting an individual path, a question about the relationship between the source node and the target node, and the question. May include an answer to.
  • the learning unit 330 inputs and encodes the source node, target node, and the relationship between the source node and the target node of each of at least one link into Long A Short-Term Memory (LSTM), and the encoded
  • LSTM Long A Short-Term Memory
  • a first result vector is generated, and the first result vector is summed in element units to form a relation network model. Secondary learning can be done.
  • At least one link may be less than or equal to a predetermined threshold number.
  • the path extraction unit 310 may extract a plurality of paths based on a path ranking algorithm (PRA).
  • PRA path ranking algorithm

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Software Systems (AREA)
  • Artificial Intelligence (AREA)
  • Computational Linguistics (AREA)
  • Evolutionary Computation (AREA)
  • Computing Systems (AREA)
  • Mathematical Physics (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
  • Machine Translation (AREA)

Abstract

La divulgation concerne un procédé permettant d'enrichir la connaissance sur la base d'un réseau de relations. Un procédé permettant d'enrichir la connaissance sur la base d'un réseau de relations selon un mode de réalisation de la présente invention comprend : une étape consistant à extraire, pour une pluralité de paires de nœuds comprises dans un graphe de connaissances, des informations de chemin qui sont des informations concernant une pluralité de chemins représentant une relation entre un nœud source et un nœud cible constituant les paires de nœuds ; une étape consistant à générer des données d'apprentissage correspondant à chacun de la pluralité de chemins sur la base des informations de chemin ; et une étape consistant à former un modèle de réseau de relations en utilisant les données d'apprentissage.
PCT/KR2020/006239 2019-11-15 2020-05-12 Procédé et dispositif permettant d'enrichir la connaissance sur la base d'un réseau de relations WO2021096009A1 (fr)

Applications Claiming Priority (4)

Application Number Priority Date Filing Date Title
KR20190146737 2019-11-15
KR10-2019-0146737 2019-11-15
KR10-2019-0163590 2019-12-10
KR1020190163590A KR102234850B1 (ko) 2019-11-15 2019-12-10 릴레이션 네트워크에 기반한 지식 보완 방법 및 장치

Publications (1)

Publication Number Publication Date
WO2021096009A1 true WO2021096009A1 (fr) 2021-05-20

Family

ID=75466363

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/KR2020/006239 WO2021096009A1 (fr) 2019-11-15 2020-05-12 Procédé et dispositif permettant d'enrichir la connaissance sur la base d'un réseau de relations

Country Status (2)

Country Link
KR (1) KR102234850B1 (fr)
WO (1) WO2021096009A1 (fr)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113297500A (zh) * 2021-06-23 2021-08-24 哈尔滨工程大学 一种社交网络孤立节点链接预测方法
CN113377955A (zh) * 2021-06-11 2021-09-10 支付宝(杭州)信息技术有限公司 文本风险发现方法和系统
CN113672740A (zh) * 2021-08-04 2021-11-19 支付宝(杭州)信息技术有限公司 针对关系网络的数据处理方法及装置
CN115391553A (zh) * 2022-08-23 2022-11-25 西北工业大学 一种自动搜索时序知识图谱补全模型的方法
CN118171727A (zh) * 2024-05-16 2024-06-11 神思电子技术股份有限公司 三元组的生成方法、装置、设备、介质及程序产品

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113360664B (zh) * 2021-05-31 2022-03-25 电子科技大学 一种知识图谱补全方法
CN113641829B (zh) * 2021-07-13 2023-11-24 北京百度网讯科技有限公司 图神经网络的训练与知识图谱的补全方法、装置
CN113672741B (zh) * 2021-08-19 2024-06-21 支付宝(杭州)信息技术有限公司 一种信息的处理方法、装置及设备
CN115525773B (zh) * 2022-10-10 2024-08-02 北京智源人工智能研究院 知识图谱补全模型的训练方法和装置
CN117575007B (zh) * 2024-01-17 2024-04-05 清华大学 基于后解码可信增强的大模型知识补全方法及系统
CN117993497A (zh) * 2024-03-15 2024-05-07 广州大学 一种基于元关系学习的知识图谱补全方法

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20130226846A1 (en) * 2012-02-24 2013-08-29 Ming Li System and Method for Universal Translating From Natural Language Questions to Structured Queries
KR20180041200A (ko) * 2015-08-19 2018-04-23 알리바바 그룹 홀딩 리미티드 정보 처리 방법 및 장치
US20180239763A1 (en) * 2017-02-17 2018-08-23 Kyndi, Inc. Method and apparatus of ranking linked network nodes
KR20190092043A (ko) * 2018-01-30 2019-08-07 연세대학교 산학협력단 추론 과정 설명이 가능한 시각 질의 응답 장치 및 방법

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20160064826A (ko) 2014-11-28 2016-06-08 한국전자통신연구원 지식 그래프 기반에서의 의미적 검색 서비스 제공장치 및 그 방법
US10681061B2 (en) * 2017-06-14 2020-06-09 International Business Machines Corporation Feedback-based prioritized cognitive analysis
US11853903B2 (en) * 2017-09-28 2023-12-26 Siemens Aktiengesellschaft SGCNN: structural graph convolutional neural network

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20130226846A1 (en) * 2012-02-24 2013-08-29 Ming Li System and Method for Universal Translating From Natural Language Questions to Structured Queries
KR20180041200A (ko) * 2015-08-19 2018-04-23 알리바바 그룹 홀딩 리미티드 정보 처리 방법 및 장치
US20180239763A1 (en) * 2017-02-17 2018-08-23 Kyndi, Inc. Method and apparatus of ranking linked network nodes
KR20190092043A (ko) * 2018-01-30 2019-08-07 연세대학교 산학협력단 추론 과정 설명이 가능한 시각 질의 응답 장치 및 방법

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
SANTORO ADAM, RAPOSO DAVID, BARRETT DAVID, MALINOWSKI MATEUSZ, PASCANU RAZVAN, BATTAGLIA PETER, LILLICRAP TIMOTHY: "A simple neural network module for relational reasoning", ARXIV.ORG, 5 June 2017 (2017-06-05), pages 1 - 16, XP080767624 *

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113377955A (zh) * 2021-06-11 2021-09-10 支付宝(杭州)信息技术有限公司 文本风险发现方法和系统
CN113297500A (zh) * 2021-06-23 2021-08-24 哈尔滨工程大学 一种社交网络孤立节点链接预测方法
CN113672740A (zh) * 2021-08-04 2021-11-19 支付宝(杭州)信息技术有限公司 针对关系网络的数据处理方法及装置
CN113672740B (zh) * 2021-08-04 2023-11-07 支付宝(杭州)信息技术有限公司 针对关系网络的数据处理方法及装置
CN115391553A (zh) * 2022-08-23 2022-11-25 西北工业大学 一种自动搜索时序知识图谱补全模型的方法
CN115391553B (zh) * 2022-08-23 2023-10-13 西北工业大学 一种自动搜索时序知识图谱补全模型的方法
CN118171727A (zh) * 2024-05-16 2024-06-11 神思电子技术股份有限公司 三元组的生成方法、装置、设备、介质及程序产品

Also Published As

Publication number Publication date
KR102234850B1 (ko) 2021-04-02

Similar Documents

Publication Publication Date Title
WO2021096009A1 (fr) Procédé et dispositif permettant d'enrichir la connaissance sur la base d'un réseau de relations
Kollias et al. Distribution matching for heterogeneous multi-task learning: a large-scale face study
WO2020122456A1 (fr) Système et procédé pour assortir de similitudes entre des images et des textes
WO2021251690A1 (fr) Système de recommandation de contenu d'apprentissage fondé sur l'entraînement de l'intelligence artificielle, et son procédé de fonctionnement
CN112528676B (zh) 文档级别的事件论元抽取方法
WO2021054514A1 (fr) Système de questions-réponses personnalisées par l'utilisateur basé sur un graphe de connaissances
WO2021095987A1 (fr) Procédé et appareil de complémentation de connaissances basée sur une entité de type multiple
WO2018212584A2 (fr) Procédé et appareil de classification de catégorie à laquelle une phrase appartient à l'aide d'un réseau neuronal profond
CN111860193B (zh) 一种基于文本的行人检索自监督视觉表示学习系统及方法
WO2013157705A1 (fr) Procédé de déduction d'un centre d'intérêt d'un utilisateur par l'intermédiaire des centres d'intérêt de voisins sociaux et de thèmes d'activités sociales dans des sns, et système à cet effet
CN110457523B (zh) 封面图片的选取方法、模型的训练方法、装置及介质
WO2022108206A1 (fr) Procédé et appareil pour remplir un graphe de connaissances pouvant être décrit
CN113011172A (zh) 文本处理方法、装置、计算机设备和存储介质
CN110322959A (zh) 一种基于知识的深度医疗问题路由方法及系统
WO2019107625A1 (fr) Procédé de traduction automatique et appareil associé
Cui et al. DGEKT: a dual graph ensemble learning method for knowledge tracing
WO2022186539A1 (fr) Procédé et appareil d'identification de célébrité basés sur la classification d'image
Zheng et al. Dynamic relevance graph network for knowledge-aware question answering
CN111522926A (zh) 文本匹配方法、装置、服务器和存储介质
Zhang et al. TS-GCN: Aspect-level sentiment classification model for consumer reviews
CN113255701B (zh) 一种基于绝对-相对学习架构的小样本学习方法和系统
CN112989024B (zh) 文本内容的关系提取方法、装置、设备及存储介质
CN113761270A (zh) 视频召回方法、装置、电子设备以及存储介质
WO2023106870A1 (fr) Système de réseau neuronal convolutif restructuré utilisant un cmp et son procédé de fonctionnement
CN116881416A (zh) 关系推理与跨模态独立匹配网络的实例级跨模态检索方法

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20888351

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 08.09.2022)

122 Ep: pct application non-entry in european phase

Ref document number: 20888351

Country of ref document: EP

Kind code of ref document: A1