利索-全球专利检索

  1. 登录
  2. 注册

搜索结果

288 条记录 (耗时 0.028 秒)

    VERIFYING NEURAL NETWORKS
    3.
    发明申请
    VERIFYING NEURAL NETWORKS 审中-公开

    公开(公告)号：WO2022243570A1

    公开(公告)日：2022-11-24

    申请号：PCT/EP2022/063919

    申请日：2022-05-23

    申请人： IMPERIAL COLLEGE INNOVATIONS LIMITED

    发明人： BATTEN, Ben ,  KOUVAROS, Panagiotis ,  LAN, Jianglin ,  LOMUSCIO, Alessio ,  ZHANG, Yang

    IPC分类号： G06N3/04 ,  G06N3/08

    摘要： There are provided processes for verifying the performance of neural networks across a range of inputs. The neural network comprises nodes arranged in a plurality of layers, and a disclosed process comprises the steps of: obtaining data representing a trained neural network, a set of algebraic constraints on the output of each layer the network, and a range of inputs to the neural network over which the algebraic constraints are to be verified, such that the data defines a verification problem; determining, for each layer of the network, a semidefinite constraint from the algebraic constraints for that layer; determining a set of interlayer constraints which constrain outputs of one or more of the layers to corresponding inputs of one or more adjacent layers; applying a semidefinite programming relaxation subject to the semidefinite constraints and the interlayer constraints across the range of inputs; and based on the outcome of the semidefinite programming relaxation, determining whether the neural network is robust across the range of inputs.

    RNA CONSTRUCT
    4.
    发明申请
    RNA CONSTRUCT 审中-公开

    公开(公告)号：WO2022129945A1

    公开(公告)日：2022-06-23

    申请号：PCT/GB2021/053362

    申请日：2021-12-17

    申请人： IMPERIAL COLLEGE INNOVATIONS LIMITED ,  VAXEQUITY LTD

    发明人： SHATTOCK, Robin ,  MCKAY, Paul ,  WATSON, Michael ,  HARPER, Elaine

    IPC分类号： A61K39/00

    摘要： The invention relates to RNA constructs, and particularly, although not exclusively, to mRNA constructs and saRNA replicons and to nucleic acids and expression vectors encoding such RNA constructs. The invention extends to the use of such RNA constructs in therapy, for example in treating diseases and/or in vaccine delivery. The invention extends to pharmaceutical compositions comprising such RNA constructs, and methods and uses thereof.

    RNA VACCINE
    5.
    发明申请
    RNA VACCINE 审中-公开

    公开(公告)号：WO2022129918A1

    公开(公告)日：2022-06-23

    申请号：PCT/GB2021/053325

    申请日：2021-12-16

    申请人： IMPERIAL COLLEGE INNOVATIONS LIMITED

    发明人： SHATTOCK, Robin

    IPC分类号： A61K39/12 ,  A61P31/14 ,  A61K39/00

    摘要： The invention relates to RNA vaccines, and particularly, although not exclusively, to novel uses and methods for thermally stabilising RNA vaccine formulations, and especially the thermal stabilisation of self-amplifying RNA vaccine formulations. The invention extends to novel vaccine compositions and formulations of temperature stabilised RNA vaccines, and their use in therapy, for example in treating and preventing disease, such as a bacterial or viral infection, and/or in vaccine delivery. The invention also extends to vaccine vials and pre-loaded syringes comprising the novel, thermally stabilised RNA vaccine formulations.

    VERIFYING NEURAL NETWORKS
    6.
    发明申请
    VERIFYING NEURAL NETWORKS 审中-公开

    公开(公告)号：WO2022117760A1

    公开(公告)日：2022-06-09

    申请号：PCT/EP2021/084042

    申请日：2021-12-02

    申请人： IMPERIAL COLLEGE INNOVATIONS LIMITED

    发明人： LOMUSCIO, Alessio ,  HENRIKSEN, Patrick ,  KOUVAROS, Panagiotis

    IPC分类号： G06N3/04 ,  G06N3/08 ,  G06N5/00

    摘要： Systems and methods are provided for verifying the transformational robustness of a neural network. Data is obtained representing a trained neural network, a set of algebraic constraints on the output of the network, and a range of inputs to the neural network over which the algebraic constraints are to be verified, such that the data defines a transformational robustness verification problem. A set of complementary constrains on the pre-activation of a node in the network are then determined such that for any input in the range of inputs, at least one of the complementary constraints is satisfied. A plurality of child verification problems are generated based on the transformational robustness verification problem and the set of complementary constraints. For each child verification problem, it is determined whether a counter-example to the child verification problem exists. Based on the determination of whether counter-examples to the child verification problems exist, it is determined whether the neural network is transformationally robust.