会员体验
专利管家(专利管理)
工作空间(专利管理)
风险监控(情报监控)
数据分析(专利分析)
侵权分析(诉讼无效)
联系我们
交流群
官方交流:
QQ群: 891211   
微信请扫码    >>>
现在联系顾问~
热词
    • 1. 发明授权
    • Monte-Carlo planning using contextual information
    • 蒙特卡洛计划使用上下文信息
    • US09047423B2
    • 2015-06-02
    • US13348993
    • 2012-01-12
    • Gerald J. TesauroAlina BeygelzimerRichard B. SegalMark N. Wegman
    • Gerald J. TesauroAlina BeygelzimerRichard B. SegalMark N. Wegman
    • G06F17/50
    • G06F17/5018G06F17/11G06N99/005G06Q10/063
    • A method, system and computer program product for choosing actions in a state of a planning problem. The system simulates one or more sequences of actions, state transitions and rewards starting from the current state of the planning problem. During the simulation of performing a given action in a given state, a data record is maintained of observed contextual state information, and observed cumulative reward resulting from the action. The system performs a regression fit on the data records, enabling estimation of expected reward as a function of contextual state. The estimations of expected rewards are used to guide the choice of actions during the simulations. Upon completion of all simulations, the top-level action which obtained highest mean reward during the simulations is recommended to be executed in the current state of the planning problem.
    • 用于在规划问题状态中选择动作的方法,系统和计算机程序产品。 系统模拟从规划问题的当前状态开始的一个或多个动作序列,状态转换和奖励。 在给定状态下执行给定动作的模拟期间,维持观察到的情境状态信息的数据记录,并且观察到由该动作产生的累积奖励。 系统对数据记录进行回归拟合,使得可以将预期报酬作为情境状态的函数进行估计。 预期奖励的估计用于指导模拟期间的动作选择。 在完成所有模拟之后,建议在计划问题的当前状态下执行在模拟期间获得最高平均奖励的顶级动作。
    • 2. 发明申请
    • MONTE-CARLO PLANNING USING CONTEXTUAL INFORMATION
    • 使用上下文信息进行蒙特卡洛规划
    • US20130185039A1
    • 2013-07-18
    • US13348993
    • 2012-01-12
    • Gerald J. TesauroAlina BeygelzimerRichard B. SegalMark N. Wegman
    • Gerald J. TesauroAlina BeygelzimerRichard B. SegalMark N. Wegman
    • G06G7/48
    • G06F17/5018G06F17/11G06N99/005G06Q10/063
    • A method, system and computer program product for choosing actions in a state of a planning problem. The system simulates one or more sequences of actions, state transitions and rewards starting from the current state of the planning problem. During the simulation of performing a given action in a given state, a data record is maintained of observed contextual state information, and observed cumulative reward resulting from the action. The system performs a regression fit on the data records, enabling estimation of expected reward as a function of contextual state. The estimations of expected rewards are used to guide the choice of actions during the simulations. Upon completion of all simulations, the top-level action which obtained highest mean reward during the simulations is recommended to be executed in the current state of the planning problem.
    • 用于在规划问题状态中选择动作的方法,系统和计算机程序产品。 系统模拟从规划问题的当前状态开始的一个或多个动作序列,状态转换和奖励。 在给定状态下执行给定动作的模拟期间,维持观察到的情境状态信息的数据记录,并且观察到由该动作产生的累积奖励。 系统对数据记录进行回归拟合,使得可以将预期报酬作为情境状态的函数进行估计。 预期奖励的估计用于指导模拟期间的动作选择。 在完成所有模拟之后,建议在计划问题的当前状态下执行在模拟期间获得最高平均奖励的顶级动作。
    • 4. 发明申请
    • METHOD FOR PERFORMANCE BOTTLENECK DIAGNOSIS AND DEPENDENCY DISCOVERY IN DISTRIBUTED SYSTEMS AND COMPUTER NETWORKS
    • 分布式系统和计算机网络中性能鉴定诊断和依赖性发现的方法
    • US20090070280A1
    • 2009-03-12
    • US11853986
    • 2007-09-12
    • Alina BeygelzimerGaurav ChandaliaIrina Rish
    • Alina BeygelzimerGaurav ChandaliaIrina Rish
    • G06N3/12
    • G06F11/3452G06F11/3495G06F2201/81G06F2201/87H04L41/5009H04L41/5038H04L43/0852H04L43/50
    • A method for performance bottleneck diagnosis and dependency discovery in distributed systems and computer networks includes receiving a real-valued end-to-end measurement matrix, a number of end-to-end measurements, a number of time points, a number of network components, a loss function, and a plurality of constraints on output matrices. The method further includes learning basic output matrices by best approximation of a transpose of the real-valued end-to-end measurement matrix, selecting a first threshold based on a real-valued mixing-weights matrix and a second threshold based on a real-valued delay matrix, converting the real-valued mixing-weights matrix and the real-valued delay matrix into respective binary matrices using the first threshold and the second threshold, and returning the real-valued mixing-weights matrix, the real-valued delay matrix, and the respective binary matrices to represent bottlenecks dependencies of the distributed system or computer network.
    • 在分布式系统和计算机网络中的性能瓶颈诊断和依赖性发现的方法包括:接收实际端到端测量矩阵,多个端对端测量,多个时间点,多个网络组件 ,损失函数,以及对输出矩阵的多个约束。 该方法还包括通过对实值端对端测量矩阵的转置进行最佳逼近来学习基本输出矩阵,基于实值混合权重矩阵和基于实数的混合权重矩阵的第二阈值来选择第一阈值, 使用第一阈值和第二阈值将实值混合权重矩阵和实值延迟矩阵转换成相应的二进制矩阵,并返回实值混合权重矩阵,实值延迟矩阵 ,以及各自的二进制矩阵来表示分布式系统或计算机网络的瓶颈依赖性。
    • 5. 发明申请
    • METHOD AND APPARATUS FOR NETWORK TOPOLOGY DISCOVERY USING CLOSURE APPROACH
    • 使用闭路方法发现网络拓扑的方法和装置
    • US20080232262A1
    • 2008-09-25
    • US11687750
    • 2007-03-19
    • Alina BeygelzimerShang Q. GuoSheng MaNatalia OdintsovaIrina Rish
    • Alina BeygelzimerShang Q. GuoSheng MaNatalia OdintsovaIrina Rish
    • H04L1/00
    • H04L41/12H04L41/0213
    • Improved network topology discovery techniques are disclosed. For example, an automated method for discovering a topology of a network, having a plurality of nodes, includes the following steps. A first traceroute process is performed for a given source node and one or more destination nodes listed in a target list so as to discover one or more paths through one or more intermediate nodes between the given source node and the one or more destination nodes. The one or more intermediate nodes discovered during the first traceroute process are added to the target list. A second traceroute process is performed for the given source node and the one or more intermediate nodes discovered during the first traceroute process so as to discover one or more paths through one or more additional intermediate nodes between the given source node and the one or more intermediate nodes discovered during the first traceroute process. When one or more additional intermediate nodes are discovered as a result of performance of the second traceroute process, the one or more additional intermediate nodes are added to the target list and the traceroute process is repeated to determine the existence of any further intermediate nodes. When no new intermediate nodes are discovered as a result of performance of the second traceroute process or a subsequent traceroute process, the nodes in the target list are identified as a target closure set for the given source node.
    • 公开了改进的网络拓扑发现技术。 例如,用于发现具有多个节点的网络拓扑的自动化方法包括以下步骤。 对给定源节点和目标列表中列出的一个或多个目的地节点执行第一跟踪路由进程,以便发现通过给定源节点和一个或多个目的地节点之间的一个或多个中间节点的一个或多个路径。 在第一个跟踪路由过程中发现的一个或多个中间节点被添加到目标列表中。 对于给定源节点和在第一跟踪路由过程期间发现的一个或多个中间节点执行第二跟踪路由进程,以便发现通过给定源节点与一个或多个中间体之间的一个或多个附加中间节点的一个或多个路径 在第一个跟踪路由过程中发现的节点。 当作为执行第二跟踪路由进程的结果发现一个或多个附加中间节点时,将一个或多个附加中间节点添加到目标列表中,并且重复跟踪路由进程以确定任何其他中间节点的存在。 由于执行第二个跟踪路由进程或后续的跟踪路由进程,没有新的中间节点被发现,目标列表中的节点被标识为给定源节点的目标封闭集。
    • 6. 发明申请
    • METHOD AND APPARATUS FOR AUTHORING AND OPTIMIZING FLOWCHARTS
    • 用于执行和优化流量的方法和装置
    • US20080227066A1
    • 2008-09-18
    • US11686959
    • 2007-03-15
    • Alina BeygelzimerMark BrodieSheng MaJonathan LenchnerIrina Rish
    • Alina BeygelzimerMark BrodieSheng MaJonathan LenchnerIrina Rish
    • G09B19/00
    • G09B7/00
    • A method and system for generating problem resolution flowcharts, whereby users do not author flowcharts directly but instead author a dependency matrix of questions and answers related by state or underlying problem cause. After creation of a matrix of questions and answers, a corresponding flowchart is then calculated based on the information in the dependency matrix, and also based on the likelihood of the various problems and their causes. The probabilities of problems and their causes may be estimated or may be calculated from historical data accumulated by use of the flow chart, or some combination of the two. These probabilities are incorporated into the answer cells of the dependency matrix. The resulting flowchart is tested and evaluated, and the results of testing and evaluation are used to make modifications in the matrix of questions and answers. The flowchart is then made available for execution by a problem determination engine, usable by staff at a help center or directly as self-help to a customer over a network. Experience from use of the problem determination engine is accumulated for use in further refinements of the dependency matrix and its probabilities.
    • 用于生成问题解决流程图的方法和系统,其中用户不直接创作流程图,而是创建与状态或潜在问题原因相关的问题和答案的依赖矩阵。 在创建了问题和答案矩阵之后,根据依赖矩阵中的信息,并根据各种问题及其原因的可能性,计算出相应的流程图。 问题的可能性及其原因可以估计,或者可以通过使用流程图累积的历史数据或两者的某种组合来计算。 这些概率被并入依赖矩阵的答案单元格中。 得到的流程图进行了测试和评估,测试和评估的结果用于对问题和答案矩阵进行修改。 然后,该流程图可由问题确定引擎执行,可由帮助中心的工作人员使用,或直接作为对网络上的客户的自助。 积累了使用问题确定引擎的经验,用于进一步细化依赖矩阵及其概率。
    • 8. 发明授权
    • Method and apparatus for authoring and optimizing flowcharts
    • 制作和优化流程图的方法和装置
    • US08341610B2
    • 2012-12-25
    • US11686959
    • 2007-03-15
    • Alina BeygelzimerMark BrodieSheng MaJonathan LenchnerIrina Rish
    • Alina BeygelzimerMark BrodieSheng MaJonathan LenchnerIrina Rish
    • G06F9/45G06F9/44
    • G09B7/00
    • A method and system for generating problem resolution flowcharts, whereby users do not author flowcharts directly but instead author a dependency matrix of questions and answers related by state or underlying problem cause. After creation of a matrix of questions and answers, a corresponding flowchart is then calculated based on the information in the dependency matrix, and also based on the likelihood of the various problems and their causes. The probabilities of problems and their causes may be estimated or may be calculated from historical data accumulated by use of the flow chart, or some combination of the two. These probabilities are incorporated into the answer cells of the dependency matrix. The resulting flowchart is tested and evaluated, and the results of testing and evaluation are used to make modifications in the matrix of questions and answers. The flowchart is then made available for execution by a problem determination engine, usable by staff at a help center or directly as self-help to a customer over a network. Experience from use of the problem determination engine is accumulated for use in further refinements of the dependency matrix and its probabilities.
    • 用于生成问题解决流程图的方法和系统,其中用户不直接创作流程图,而是创建与状态或潜在问题原因相关的问题和答案的依赖矩阵。 在创建了问题和答案矩阵之后,根据依赖矩阵中的信息,并根据各种问题及其原因的可能性,计算出相应的流程图。 问题的可能性及其原因可以估计,或者可以通过使用流程图累积的历史数据或两者的某种组合来计算。 这些概率被并入依赖矩阵的答案单元格中。 所得流程图经过测试和评估,测试和评估结果用于对问题和答案矩阵进行修改。 然后,该流程图可由问题确定引擎执行,可由帮助中心的工作人员使用,或直接作为对网络上的客户的自助。 积累了使用问题确定引擎的经验,用于进一步细化依赖矩阵及其概率。
    • 10. 发明申请
    • Method and System for Deducing Network Routes by Querying Routers
    • 通过查询路由器来减少网络路由的方法和系统
    • US20100091685A1
    • 2010-04-15
    • US12251094
    • 2008-10-14
    • Dakshi AgrawalAlina BeygelzimerBong-Jun KoMudhakar Srivatsa
    • Dakshi AgrawalAlina BeygelzimerBong-Jun KoMudhakar Srivatsa
    • H04L12/28
    • H04L45/22H04L45/02H04L45/021H04L45/28H04L45/54
    • A method and system for detecting and correcting invalid network paths in a network having a plurality of routers and a set of global invariant rules imbedded in the network including querying a subset of a routing table data structure defining network paths wherein said subset includes a plurality of routing table entries and comparing a network routing table entry with an earlier network routing table entry for the same node. The method and system also includes detecting inconsistencies between the newer network routing table entry and the earlier network routing table entry, and invalidating the network entries of the older network routing table that are inconsistent with the newer entries. The method and system also includes reconstructing the routing structure by replacing the invalidated network routing table entries with new routing table entries that minimize violation of the invariant protocols.
    • 一种用于检测和校正具有嵌入在所述网络中的多个路由器和一组全局不变规则的网络中的无效网络路径的方法和系统,包括查询定义网络路径的路由表数据结构的子集,其中所述子集包括多个 路由表条目,并将网络路由表条目与同一节点的早期网络路由表条目进行比较。 该方法和系统还包括检测较新的网络路由表项和较早的网络路由表项之间的不一致性,并使旧网络路由表的与较新条目不一致的网络条目失效。 该方法和系统还包括通过将无效网络路由表条目替换为最小化违反不变协议的新路由表条目来重建路由结构。