知搜-全球专利检索

  1. Login
  2. Sign Up

Search Results

24 records (took 0.045 seconds)

    Statistical Design with Importance Sampling Reuse
    1.
    发明申请
    Statistical Design with Importance Sampling Reuse 审中-公开
    统计设计与重要性抽样重用

    公开(公告)号：US20160266950A1

    公开(公告)日：2016-09-15

    申请号：US15161462

    申请日：2016-05-23

    Applicant: International Business Machines Corporation

    Inventor： Rajiv V. Joshi ,  Rouwaida N. Kanj ,  Sani R. Nassif ,  Carl J. Radens

    IPC: G06F11/07

    CPC classification number: G06F11/076 ,  G06F11/008 ,  G06F11/079 ,  G06F17/5009 ,  G06F17/5022 ,  G06F2217/10

    Abstract: A mechanism is provided for reusing importance sampling for efficient cell failure rate estimation of process variations and other design considerations. First, the mechanism performs a search across circuit parameters to determine failures with respect to a set of performance variables. For a single failure region, the initial search may be a uniform sampling of the parameter space. Mixture importance sampling (MIS) efficiently may estimate the single failure region. The mechanism then finds a center of gravity for each metric and finds importance samples. Then, for each new origin corresponding to a process variation or other design consideration, the mechanism finds a suitable projection and recomputes new importance sampling (IS) ratios.

    Abstract translation: 提供了一种用于重用采样的机制，用于有效地进行细胞故障率估计过程变化和其他设计考虑。 首先，该机制对电路参数进行搜索，以确定相对于一组性能变量的故障。 对于单个故障区域，初始搜索可以是参数空间的均匀采样。 混合重要性抽样（MIS）有效地估计单个故障区域。 然后，该机制找到每个度量的重心，并发现重要性样本。 然后，对于对应于过程变化或其他设计考虑的每个新的原点，机制找到合适的投影并重新计算新的重要性抽样（IS）比率。

    SELECTIVE IMPORTANCE SAMPLING
    2.
    发明申请
    SELECTIVE IMPORTANCE SAMPLING 有权
    选择性重要采样

    公开(公告)号：US20140278309A1

    公开(公告)日：2014-09-18

    申请号：US13837900

    申请日：2013-03-15

    Applicant: INTERNATIONAL BUSINESS MACHINES CORPORATION

    Inventor： Rajiv V. Joshi ,  Rouwaida N. Kanj ,  Pranita Kerber

    IPC: G06F17/50

    CPC classification number: G06F17/5009

    Abstract: The present disclosure relates generally to the field of selective importance sampling. In various examples, selective importance sampling may be implemented in the form of systems and/or algorithms.

    Abstract translation: 本公开一般涉及选择性重要性抽样的领域。 在各种示例中，可以以系统和/或算法的形式实现选择性重要性抽样。

    Statistical design with importance sampling reuse
    3.
    发明授权
    Statistical design with importance sampling reuse 有权

    公开(公告)号：US11372701B2

    公开(公告)日：2022-06-28

    申请号：US16543776

    申请日：2019-08-19

    Applicant: International Business Machines Corporation

    Inventor： Rajiv V. Joshi ,  Rouwaida N. Kanj ,  Sani R. Nassif ,  Carl J. Radens

    IPC: G06F30/33 ,  G06F30/20 ,  G06F111/08 ,  G06F11/07 ,  G06F11/00

    Abstract: A mechanism is provided for reusing importance sampling for efficient cell failure rate estimation of process variations and other design considerations. First, the mechanism performs a search across circuit parameters to determine failures with respect to a set of performance variables. For a single failure region, the initial search may be a uniform sampling of the parameter space. Mixture importance sampling (MIS) efficiently may estimate the single failure region. The mechanism then finds a center of gravity for each metric and finds importance samples. Then, for each new origin corresponding to a process variation or other design consideration, the mechanism finds a suitable projection and recomputes new importance sampling (IS) ratios.

    Statistical design with importance sampling reuse
    6.
    发明授权
    Statistical design with importance sampling reuse 有权
    具有重要性抽样重用的统计设计

    公开(公告)号：US09348680B2

    公开(公告)日：2016-05-24

    申请号：US14242418

    申请日：2014-04-01

    Applicant: International Business Machines Corporation

    Inventor： Rajiv V. Joshi ,  Rouwaida N. Kanj ,  Sani R. Nassif ,  Carl J. Radens

    IPC: G06F11/00 ,  G06F17/50

    CPC classification number: G06F11/076 ,  G06F11/008 ,  G06F11/079 ,  G06F17/5009 ,  G06F17/5022 ,  G06F2217/10

    Abstract: A mechanism is provided for reusing importance sampling for efficient cell failure rate estimation of process variations and other design considerations. First, the mechanism performs a search across circuit parameters to determine failures with respect to a set of performance variables. For a single failure region, the initial search may be a uniform sampling of the parameter space. Mixture importance sampling (MIS) efficiently may estimate the single failure region. The mechanism then finds a center of gravity for each metric and finds importance samples. Then, for each new origin corresponding to a process variation or other design consideration, the mechanism finds a suitable projection and recomputes new importance sampling (IS) ratios.

    Abstract translation: 提供了一种用于重用采样的机制，用于有效地进行细胞故障率估计过程变化和其他设计考虑。 首先，该机制对电路参数进行搜索，以确定相对于一组性能变量的故障。 对于单个故障区域，初始搜索可以是参数空间的均匀采样。 混合重要性抽样（MIS）有效地估计单个故障区域。 然后，该机制找到每个度量的重心，并发现重要性样本。 然后，对于对应于过程变化或其他设计考虑的每个新的原点，机制找到合适的投影并重新计算新的重要性抽样（IS）比率。

    DOCUMENT TAGGING AND RETRIEVAL USING PER-SUBJECT DICTIONARIES INCLUDING SUBJECT-DETERMINING-POWER SCORES FOR ENTRIES
    7.
    发明申请
    DOCUMENT TAGGING AND RETRIEVAL USING PER-SUBJECT DICTIONARIES INCLUDING SUBJECT-DETERMINING-POWER SCORES FOR ENTRIES 有权
    使用包含用于入门的主体确定功率的主题词语的文档标记和检索

    公开(公告)号：US20160041975A1

    公开(公告)日：2016-02-11

    申请号：US14887495

    申请日：2015-10-20

    Applicant: INTERNATIONAL BUSINESS MACHINES CORPORATION

    Inventor： Anne Elizabeth Gattiker ,  Fadi H. Gebara ,  Anthony N. Hylick ,  Rouwaida N. Kanj ,  Jian Li

    IPC: G06F17/30 ,  G06F17/27

    CPC classification number: G06F17/30616 ,  G06F17/2735 ,  G06F17/30011 ,  G06F17/30424

    Abstract: Techniques for managing big data include tagging of documents and subsequent retrieval using per-subject dictionaries having entries with subject-determining-power scores. The subject-determining-power scores provide an indication of the descriptive power of the term with respect to the subject of the dictionary containing the term. The same term may have entries in multiple dictionaries with different subject-determining-power scores in each of the dictionaries. A retrieval request for one or more documents containing search terms descriptive of the one or more documents can be processed identifying a set of candidate documents tagged with subjects and optional terms, and then applying subject-determining-power scores from the multiple dictionaries for the search term to determine a subject for the search term. The method then selects the one or more documents from the candidate documents according to the subject.

    Abstract translation: 用于管理大数据的技术包括使用具有主题确定能力分数的条目的每个主题词典来标记文档和随后的检索。 受试者决定力分数提供关于包含该术语的词典的主题的术语的描述力的指示。 同一术语可能在多个词典中具有不同的词典确定能力得分的条目。 可以对包含描述一个或多个文档的搜索项的一个或多个文档的检索请求进行处理，以识别标记有主题和可选项的候选文本集合，然后从多个字典中应用主题确定权力分数进行搜索 以确定搜索词的主题。 然后，该方法根据对象从候选文档中选择一个或多个文档。

    CIRCUIT FOR MEMORY CELL RECOVERY
    10.
    发明申请
    CIRCUIT FOR MEMORY CELL RECOVERY 失效
    用于记忆细胞恢复的电路

    公开(公告)号：US20130272077A1

    公开(公告)日：2013-10-17

    申请号：US13915913

    申请日：2013-06-12

    Applicant: International Business Machines Corporation

    Inventor： Rajiv V. Joshi ,  Rouwaida N. Kanj ,  Jente B. Kuang ,  Carl J. Radens

    IPC: G11C7/00

    CPC classification number: G11C7/00 ,  G11C7/02 ,  G11C7/04 ,  G11C11/417 ,  G11C11/419

    Abstract: An apparatus and method for combating the effects of bias temperature instability (BTI) and other variability in a memory cell. Bit lines connecting to a memory cell contain two alternate paths that criss-cross to connect a lower portion of a first bit line to an upper portion of a second bit line, and to connect a lower portion of the second bit line to an upper portion of the first bit line. Alternative to activating transistors on the bit lines to read and write to the memory cell, transistors on the alternative paths may be activated to read and write to the memory cell from the opposite bit line. In this fashion, the memory cell may be read through the bit lines to a sense amplifier where the bit values are latched. While the bit values remain latched in the sense amplifier, the transistors on the bit lines are deactivated and the transistors on the alternate paths are activated. When the word line is accessed, the bit values will be written into the opposite sides of the memory cell, reversing the biases.

    Abstract translation: 用于抵抗偏置温度不稳定性（BTI）和存储器单元中其他变异性的影响的装置和方法。 连接到存储单元的位线包含两个交替的交叉路径，其交叉连接第一位线的下部与第二位线的上部，并且将第二位线的下部连接到上部 的第一个位线。 在位线上激活晶体管以读取和写入存储器单元的替代方案是替代路径上的晶体管可被激活以从相对位线读取和写入存储器单元。 以这种方式，存储器单元可以通过位线读取到其中位值被锁存的读出放大器。 当位值保持锁存在读出放大器中时，位线上的晶体管被​​去激活，并且备用路径上的晶体管被​​激活。 当访问字线时，位值将写入存储单元的相对侧，反转偏置。

Patent Agency Ranking