知搜-全球专利检索

  1. 登录
  2. 注册

搜索结果

57 条记录 (耗时 0.027 秒)

    KNOWLEDGE AUGMENTED SEQUENTIAL DECISION-MAKING UNDER UNCERTAINTY
    21.
    发明公开
    KNOWLEDGE AUGMENTED SEQUENTIAL DECISION-MAKING UNDER UNCERTAINTY 审中-公开

    公开(公告)号：US20230186145A1

    公开(公告)日：2023-06-15

    申请号：US17548672

    申请日：2021-12-13

    申请人： International Business Machines Corporation

    发明人： Radu Marinescu

    IPC分类号： G06N20/00 ,  G06N7/00 ,  G06F40/40

    CPC分类号： G06N20/00 ,  G06N7/00 ,  G06F40/40

    摘要： One or more systems, devices, computer program products and/or computer-implemented methods of use provided herein relate to outputting an optimal decision policy base on informal knowledge input. A system can comprise a memory that stores computer executable components, and a processor that executes the computer executable components stored in the memory, wherein the computer executable components can comprise an analysis component that analyzes an input dataset comprising a constraint in a natural language form, and an augmentation component that generates an influence mapping comprising a constraint variable based on the constraint input. In an embodiment, an input dataset employed to support the influence mapping can comprise time-stamped tuple data comprising a state, an action and a reward. In an embodiment, an inference engine can generate an output policy in response to the constraint input and which output policy can be based on the constraint input and constraint variable.

    OPTIMIZED GROUP BUYING SCHEMES
    23.
    发明申请
    OPTIMIZED GROUP BUYING SCHEMES 有权

    公开(公告)号：US20220188902A1

    公开(公告)日：2022-06-16

    申请号：US17120320

    申请日：2020-12-14

    申请人： International Business Machines Corporation

    发明人： Radu Marinescu ,  Akihiro Kishimoto

    IPC分类号： G06Q30/06

    摘要： In an approach for generating and recommending optimized shopping orders for a group of users that collectively purchase bundles of goods, a processor generates an initial shopping order for each user in a group of shopping users, based on one or more preferences and constraints of each user on one or more items to buy from a stock. A processor optimizes the initial shopping order for each user based on one or more objectives of each user. A processor outputs the optimized shopping order for each user.

    GENERATING SYMBOLIC DOMAIN MODELS FROM MULTIMODAL DATA
    24.
    发明申请
    GENERATING SYMBOLIC DOMAIN MODELS FROM MULTIMODAL DATA 有权

    公开(公告)号：US20220100968A1

    公开(公告)日：2022-03-31

    申请号：US17035777

    申请日：2020-09-29

    申请人： International Business Machines Corporation

    发明人： Mattia Chiari ,  Yufang Hou ,  Hiroshi Kajino ,  Akihiro Kishimoto ,  Radu Marinescu

    IPC分类号： G06F40/47 ,  G06N3/08 ,  G06N3/04 ,  G06F40/51

    摘要： A computer generates a formal planning domain description. The computer receives a first text-based description of a domain in an AI environment. The domain includes an action and an associated attribute, and the description is written in natural language. The computer receives the first text-based description of said domain and extracts a first set of domain actions and associated action attributes. The computer receives audio-visual elements depicting the domain, generates a second text-based description, and extracts a second set of domain actions and associated action attributes. The computer constructs finite state machines corresponding to the extracted actions and attributes. The computer converts the FSMs into a symbolic model, written in a formal planning language, that describes the domain.

    Interactive training for application providers
    25.
    发明授权
    Interactive training for application providers 有权

    公开(公告)号：US11217116B2

    公开(公告)日：2022-01-04

    申请号：US15938197

    申请日：2018-03-28

    申请人： INTERNATIONAL BUSINESS MACHINES CORPORATION

    发明人： Radu Marinescu ,  Alice-Maria Marascu

    IPC分类号： G09B19/00 ,  G09B5/00 ,  G10L15/26 ,  G09B7/00

    摘要： A system and method for interactive training for application providers in a computing environment are presented. A proposed application solution from a user for a selected application may be compared to one or more optimized solutions to identify one or more differences in the proposed application solution. One or more missing assets may be identified from the proposed application solution according to the one or more differences. The user may be surveyed with a survey relating to the missing assets such that survey results are used to train and develop a level of expertise for the user.

    SOLVER DEVICES AND METHODS
    29.
    发明公开
    SOLVER DEVICES AND METHODS 审中-公开

    公开(公告)号：US20240362498A1

    公开(公告)日：2024-10-31

    申请号：US18309625

    申请日：2023-04-28

    申请人： INTERNATIONAL BUSINESS MACHINES CORPORATION

    发明人： Rahul Nair ,  Radu Marinescu ,  Ambrish Rawat ,  Daniel Karl I. Weidele

    IPC分类号： G06N5/01

    CPC分类号： G06N5/013

    摘要： A system includes an agent engine, an encoder, a general-purpose solver engine, and an orchestrator. The orchestrator is configured to receive a first problem instance corresponding to a learned policy that is based on auto reinforcement learning, and provide the first problem instance to the general-purpose solver engine, which is configured to execute based on the first problem instance to determine a solver state. The orchestrator is configured to extract, from the general-purpose solver engine, the solver state, and to provide the solver state to the encoder. The encoder is configured to query the agent engine for a best action according to the learned policy and an encoded solver state. The agent engine is configured to determine the best action according to the learned policy and the encoded solver state. The orchestrator is configured to receive the best action, and direct the general-purpose solver to implement the best action.

    REINFORCEMENT LEARNING WITH MULTIPLE OBJECTIVES AND TRADEOFFS
    30.
    发明公开
    REINFORCEMENT LEARNING WITH MULTIPLE OBJECTIVES AND TRADEOFFS 审中-公开

    公开(公告)号：US20240232682A9

    公开(公告)日：2024-07-11

    申请号：US17972291

    申请日：2022-10-24

    申请人： INTERNATIONAL BUSINESS MACHINES CORPORATION

    发明人： Radu Marinescu ,  Parikshit Ram ,  Djallel Bouneffouf ,  Tejaswini Pedapati ,  Paulito Palmes

    IPC分类号： G06N20/00 ,  G06F7/544

    CPC分类号： G06N20/00 ,  G06F7/5443

    摘要： A method for computing possibly optimal policies in reinforcement learning with multiple objectives and tradeoffs includes receiving a dataset comprising state, action, and reward information for objectives in a multiple objective environment. Tradeoff information indicating that a first vector comprising first values of the objectives in the multiple objective environment is preferred to a second vector comprising second values of the objectives in the multiple objective environment is received. A set of possibly optimal policies for the multiple objective environment is produced based on the dataset and the tradeoff information, where the set of possibly optimal policies indicates actions for an intelligent agent operating in the multiple objective environment to take.