中科微点-全球专利检索

  1. Login
  2. Sign Up

Search Results

6 records (took 0.016 seconds)

    Time-series forecasting and machine-learning model parameter optimization
    1.
    发明专利
    Time-series forecasting and machine-learning model parameter optimization 未知

    公开(公告)号：GB2633833A

    公开(公告)日：2025-03-26

    申请号：GB202314621

    申请日：2023-09-25

    Applicant: IBM

    Inventor： CHRISTOPHER WATSON ,  ANDRÉS GARCIA SARAVIA ORTIZ DE MONTELLANO ,  DIETER WELLERDIEK ,  JAKOB LANG ,  GIAO NGUYEN-QUYNH

    IPC: G06N20/00 ,  G06F18/211 ,  G06F18/214

    Abstract: A method for time-series forecasting for time-series data with a periodic behavior larger than a respective sampling rate comprises: selecting (104) candidate time lag values from measured (102) time-series data; determining (106) first training data, e.g. feature vectors, based on time-series data and the candidate time lag values; and training (108) a first machine-learning (ML) system for building a regularized ML model, thereby determining a subset of the set of the first training data related to the most influential long-term and short-term lag values when training the regularised ML model. The method further comprises: building (110) second training data based on the long-term lag values and related measured sampled time-series data; training (112) of a second machine-learning system for time-series predictions when using measured sampled time-series data as input, the training using the set of second training data, related measured sampled time-series, the short-term lag values; wherein a first performance indicator value is indicative of a prediction performance of the first time-series machine-learning model; determining (backward optimization 114) that an element of the set of second training data is significant for the training of the first time-series machine-learning model; and determining (forward optimisation 116), that the set of second training data is complete. A table with measured sampled time-series data shifted by candidate time lag values is used for determining (106) first training data. Determining the second training data being complete comprises a Fourier transformation of an error signal between ground truth and predicted time series.

    Integrating a communication bridge into a data procesing system
    2.
    发明专利
    Integrating a communication bridge into a data procesing system 未知

    公开(公告)号：GB2532732A

    公开(公告)日：2016-06-01

    申请号：GB201420905

    申请日：2014-11-25

    Applicant: IBM

    Inventor： RENE TRUMPP ,  JAKOB LANG ,  ANDREAS R MAIER ,  ANGEL NUNEZ MENCIAS

    IPC: H04L69/40 ,  G06F11/16

    Abstract: A further communication bridge 14 is integrated into a running data processing system 210, which comprises a communication client 10 running a first operating system 16 having no own communication stack and at least a first communication bridge 12 running a second operating system 18 having an own communication stack 20. The first communication bridge 12 acts as a master communication bridge 26 and the further communication bridge 14 runs a third operating system 19 having an own communication stack 22. The further communication bridge 14 announces itself as a slave communication bridge 28 to the master communication bridge 26 at an announcement time and the master communication bridge 26 executes a quiesce process on a network adapter 68 and on an API 62 of the communication client 10 when there are no data packets in a queue with a sending time earlier than the announcement time. The master communication bridge 26 extracts the state of its communication stack 20, sends it to the further communication bridge 14 and resumes the network adapter 68 and the API 62.

    Integrating a communication bridge into a data procesing system
    3.
    发明专利
    Integrating a communication bridge into a data procesing system 未知

    公开(公告)号：GB2532732B

    公开(公告)日：2019-06-26

    申请号：GB201420905

    申请日：2014-11-25

    Applicant: IBM

    Inventor： RENE TRUMPP ,  JAKOB LANG ,  ANDREAS R MAIER ,  ANGEL NUNEZ MENCIAS

    IPC: H04L69/40 ,  G06F11/16

    Abstract: Integrating a further communication bridge into a running data processing system. The data processing system includes a communication client running a first operating system having no own communication stack and at least a first communication bridge running a second operating system having an own communication stack. The first communication bridge is configured as a master communication bridge. The further communication bridge announces itself as a slave communication bridge at an announcement time. The master communication bridge executes a quiesce process on the network adapter and on the API of the communication client when there are no data packets in the queue with a sending time earlier than the announcement time. The master communication bridge extracts the state of its communication stack and sends it to the further communication bridge. The master communication bridge resumes the network adapter and the API.

    Dynamically assigning network addresses
    4.
    发明专利
    Dynamically assigning network addresses 未知

    公开(公告)号：GB2558163A

    公开(公告)日：2018-07-11

    申请号：GB201420129

    申请日：2014-11-12

    Applicant: IBM

    Inventor： ANGEL NUNEZ MENCIAS ,  JAKOB LANG ,  MARTIN TROESTER ,  THOMAS POHL

    IPC: H04L29/12 ,  G06F9/455

    Abstract: A server (14) is provided in a network (210) for dynamically assigning network addresses to virtual network adapters in virtual machines (VMs) (10,12). A sniffer (30) extracts network IP addresses associated with MAC addresses of the VMs from packets sent between the server and the VMs. A table (26, 28) stores the extracted addresses with the corresponding MAC addresses and status information for each VM. If a VM status is suspended, a daemon (32) periodically sends renewal requests to the server so that the VMs assigned IP address is not reassigned to another network component while it is suspended. The renewal messages cease when the VM returns to operation.

Patent Agency Ranking