公开(公告)号：US09575809B2

公开(公告)日：2017-02-21

申请号：US13347674

申请日：2012-01-10

申请人： Eyal Lubetzky ,  Yuval Peres ,  Dahlia Malkhi ,  Po-Shen Loh

发明人： Eyal Lubetzky ,  Yuval Peres ,  Dahlia Malkhi ,  Po-Shen Loh

IPC分类号： G06F15/16 ,  G06F9/50 ,  H04L12/12

CPC分类号： G06F9/5061 ,  H04L12/12 ,  Y02D50/40

摘要： A “Stochastic Clustering-Based Network Generator” enables rapid formation of an interconnected hierarchical network structure from an arbitrary number of agents via an iterative turn-based coalescence process. Given N agents wishing to coalesce into one hierarchical network, a turn-based process allows each agent (or the head of each hierarchical cluster of agents), to randomly decide whether to issue or listen for merge requests in each round. Issuing a request amounts to contacting a randomly chosen agent with a merge request. Given multiple received requests, a cluster head will randomly accept one request for a merge received by any agent in that cluster. The requesting cluster then merges as a hierarchical child of the accepting cluster. In a related embodiment, given multiple merge requests, the request from the smallest cluster is accepted. In further embodiments, ties of the smallest cluster size are broken based on various options.

摘要翻译： “随机聚类网络生成器”可以通过迭代的基于回合的聚结过程从任意数量的代理中快速形成互连的分层网络结构。 给定希望合并成一个分层网络的N个代理，基于回合的进程允许每个代理（或代理的每个分级集群的头部）随机地决定是否在每一轮中发出或收听合并请求。 发出请求等于将随机选择的代理与合并请求联系起来。 给定多个接收到的请求时，簇头将随机接受一个该集群中任何代理收到的合并请求。 然后，请求集合作为接受集群的分层子进行合并。 在相关实施例中，给定多个合并请求，来自最小簇的请求被接受。 在另外的实施例中，基于各种选项来破坏最小簇大小的关系。

公开(公告)号：US20090170584A1

公开(公告)日：2009-07-02

申请号：US11965772

申请日：2007-12-28

申请人： Desney S. Tan ,  Gregory R. Smith ,  Yuval Peres ,  Joseph Yossi Azar ,  Eyal Lubetzky

发明人： Desney S. Tan ,  Gregory R. Smith ,  Yuval Peres ,  Joseph Yossi Azar ,  Eyal Lubetzky

IPC分类号： A63F9/24

CPC分类号： G07F17/3276 ,  G07F17/32

摘要： A tournament-style gaming scenario exploration system and method for interactively exploring current and future scenarios of a tournament and associated pick'em pool. The system and method include a prediction module (including a game constraint sub-module), and a key event detection module. Embodiments of the prediction module include a binary integer that represents tournament outcomes. The prediction module generates predictions of tournament outcomes using an exhaustive or a sampling technique. The sampling technique includes random sampling, where the tournament bracket is randomly sampled, and a weighted sampling technique, which sample portions of the tournament bracket more densely than others areas. Embodiments of the game constraint sub-module allow real-world results constraints and user-supplied constraints to be imposed on the tournament outcomes. Embodiments of the key event detection module identify key games in the tournament that affect a user's placement in the pick'em pool, a competitor's placement in the tournament standings, or both.

摘要翻译： 一种比赛风格的游戏场景探索系统和方法，用于交互式探索比赛和相关选择队列的当前和未来场景。 该系统和方法包括预测模块（包括游戏约束子模块）和键事件检测模块。 预测模块的实施例包括表示赛事结果的二进制整数。 预测模块使用穷举或抽样技术来产生比赛成果的预测。 采样技术包括随机采样，其中随机采样比赛支架，以及加权采样技术，其中比赛部分比其他地区更加密集。 游戏约束子模块的实施例允许实际结果约束和用户提供的约束被施加在比赛结果上。 关键事件检测模块的实施例识别比赛中的关键游戏，影响用户在选择池中的位置，竞争对手在比赛积分中的位置，或两者兼而有之。

公开(公告)号：US20110270774A1

公开(公告)日：2011-11-03

申请号：US12772083

申请日：2010-04-30

申请人： Roy Varshavsky ,  Moshe Tennenholtz ,  Yuval Peres ,  David B. Wilson

发明人： Roy Varshavsky ,  Moshe Tennenholtz ,  Yuval Peres ,  David B. Wilson

IPC分类号： G06Q30/00 ,  G06Q99/00

CPC分类号： H04M7/0024 ,  G06Q10/10 ,  G06Q30/0282 ,  G06Q50/01 ,  H04L63/126 ,  H04L67/22 ,  H04M2203/655

摘要： Providing a recommendation to a group of networked members is disclosed. The recommendation is provided to the group collectively, and is based on trust relationships between the members of the network. In an example embodiment, the network is a social network. Example systems and methods include a two-phase approach and a one-phase approach, each including analysis and aggregation of input associated with members of the network.

摘要翻译： 披露了向一组网络成员提出的建议。 该建议集体提供给集团，并基于网络成员之间的信任关系。 在示例实施例中，网络是社交网络。 示例系统和方法包括两阶段方法和单阶段方法，每个方法包括与网络成员相关联的输入的分析和聚合。

公开(公告)号：US20100318404A1

公开(公告)日：2010-12-16

申请号：US12484279

申请日：2009-06-15

申请人： Kamal Jain ,  Yuval Peres ,  Nikhil Rangarajan

发明人： Kamal Jain ,  Yuval Peres ,  Nikhil Rangarajan

IPC分类号： G06Q30/00 ,  G06Q10/00 ,  G06Q50/00

CPC分类号： G06Q30/0214 ,  G06Q30/02 ,  G06Q30/04

摘要： Consumer stores typically sell products at a retail profit, which covers the cost of pre-sale, consumer-influencing services regarding the products (e.g., discussions with sales associates, product demonstrations, and consumer inspections of the products.) However, consumers may utilize these services but may purchase selected products from other retailers that do not provide pre-sale services, and that may accordingly offer lower prices. This tactic may cause lost sales for consumer stores, and eventually store failures. Instead, the consumer store may charge the supplier an influence fee for consumer-influencing services based on a per-service price established between the supplier and the consumer store and consumer influence metrics tracked by a computer. The supplier may recoup the influence fee in the wholesale cost of the product, thereby distributing the shared benefit of the influencing services of the consumer store across all retailers and enabling a sustainable business model for providing the consumer-influencing services.

摘要翻译： 消费品商店通常以零售利润销售产品，其中包括预售成本，对产品的消费者影响服务（例如，与销售人员的讨论，产品演示和产品的消费者检查）。但是，消费者可能利用 这些服务，但可能从不提供售前服务的其他零售商处购买选定的产品，因此可以提供较低的价格。 这种策略可能导致消费者商店的销售损失，并最终存储失败。 相反，消费者商店可以根据供应商和消费者商店之间建立的每个服务价格以及由计算机跟踪的消费者影响度量，向供应商收取对消费者影响服务的影响费。 供应商可以在产品的批发成本中收回影响费，从而在所有零售商分销消费品店的影响力服务的共同利益，并为可持续发展的商业模式提供消费者影响力的服务。

公开(公告)号：US20100205126A1

公开(公告)日：2010-08-12

申请号：US12367404

申请日：2009-02-06

申请人： Reid Marlow Andersen ,  Yuval Peres

发明人： Reid Marlow Andersen ,  Yuval Peres

IPC分类号： G06N3/12

CPC分类号： G06F17/10 ,  H04L41/145 ,  H04L41/16

摘要： Providing for local graph partitioning using an evolving set process is disclosed herein. By way of example, a computer processor can be configured to execute local partitioning based on evolving set instructions. The instructions can be employed to transition a set of analyzed vertices of a graph until a segment of the graph with small conductance is identified. A transitioning algorithm can expand or contract the analyzed set of vertices based on characteristics of vertices at a boundary of the analyzed set. Accordingly, as the set of analyzed vertices becomes large, significant processing efficiency is gained by employing the characteristics of boundary vertices to transition the set or determine conductance, rather than all vertices of the analyzed set.

摘要翻译： 本文公开了使用演进集合过程提供局部图分割。 作为示例，计算机处理器可以被配置为基于演进的设置指令来执行本地分区。 可以使用这些指令来转换图的一组经分析的顶点，直到识别出具有小电导的图形的一段。 转换算法可以基于分析集合的边界处的顶点的特征来扩展或收缩分析的顶点集合。 因此，当分析的顶点集合变大时，通过采用边界顶点的特征来转换集合或确定电导而不是分析集合的所有顶点，可以获得显着的处理效率。

公开(公告)号：US20130179487A1

公开(公告)日：2013-07-11

申请号：US13347674

申请日：2012-01-10

申请人： Eyal Lubetzky ,  Yuval Peres ,  Dahlia Malkhi ,  Po-Shen Loh

发明人： Eyal Lubetzky ,  Yuval Peres ,  Dahlia Malkhi ,  Po-Shen Loh

IPC分类号： G06F15/16

CPC分类号： G06F9/5061 ,  H04L12/12 ,  Y02D50/40

摘要： A “Stochastic Clustering-Based Network Generator” enables rapid formation of an interconnected hierarchical network structure from an arbitrary number of agents via an iterative turn-based coalescence process. Given N agents wishing to coalesce into one hierarchical network, a turn-based process allows each agent (or the head of each hierarchical cluster of agents), to randomly decide whether to issue or listen for merge requests in each round. Issuing a request amounts to contacting a randomly chosen agent with a merge request. Given multiple received requests, a cluster head will randomly accept one request for a merge received by any agent in that cluster. The requesting cluster then merges as a hierarchical child of the accepting cluster. In a related embodiment, given multiple merge requests, the request from the smallest cluster is accepted. In further embodiments, ties of the smallest cluster size are broken based on various options.

摘要翻译： “随机聚类网络生成器”可以通过迭代的基于回合的聚结过程从任意数量的代理中快速形成互连的分层网络结构。 给定希望合并成一个分层网络的N个代理，基于回合的进程允许每个代理（或代理的每个分级集群的头）随机地决定是否在每一轮中发出或收听合并请求。 发出请求等于将随机选择的代理与合并请求联系起来。 给定多个接收到的请求时，簇头将随机接受一个该集群中任何代理收到的合并请求。 然后，请求集合作为接受集群的分层子进行合并。 在相关实施例中，给定多个合并请求，来自最小簇的请求被接受。 在另外的实施例中，基于各种选项来破坏最小簇大小的关系。

公开(公告)号：US08364615B2

公开(公告)日：2013-01-29

申请号：US12367404

申请日：2009-02-06

申请人： Reid Marlow Andersen ,  Yuval Peres

发明人： Reid Marlow Andersen ,  Yuval Peres

IPC分类号： G06F15/18 ,  G06F17/50 ,  G06F7/00

CPC分类号： G06F17/10 ,  H04L41/145 ,  H04L41/16

摘要： Providing for local graph partitioning using an evolving set process is disclosed herein. By way of example, a computer processor can be configured to execute local partitioning based on evolving set instructions. The instructions can be employed to transition a set of analyzed vertices of a graph until a segment of the graph with small conductance is identified. A transitioning algorithm can expand or contract the analyzed set of vertices based on characteristics of vertices at a boundary of the analyzed set. Accordingly, as the set of analyzed vertices becomes large, significant processing efficiency is gained by employing the characteristics of boundary vertices to transition the set or determine conductance, rather than all vertices of the analyzed set.

摘要翻译： 本文公开了使用演进集合过程提供局部图分割。 作为示例，计算机处理器可以被配置为基于演进的设置指令来执行本地分区。 可以使用这些指令来转换图的一组经分析的顶点，直到识别出具有小电导的图形的一段。 转换算法可以基于分析集合的边界处的顶点的特征来扩展或收缩分析的顶点集合。 因此，当分析的顶点集合变大时，通过采用边界顶点的特征来转换集合或确定电导而不是分析集合的所有顶点，可以获得显着的处理效率。