公开(公告)号：US08788578B2

公开(公告)日：2014-07-22

申请号：US13180375

申请日：2011-07-11

Applicant: Anthony Wood ,  David Stern ,  David Westerhoff ,  Bill Ataras ,  Mark Goodwin ,  Aaron Graham ,  Matt Emerson ,  Kevin Phillips

Inventor： Anthony Wood ,  David Stern ,  David Westerhoff ,  Bill Ataras ,  Mark Goodwin ,  Aaron Graham ,  Matt Emerson ,  Kevin Phillips

IPC: G06F15/16

CPC classification number: H04L65/4069 ,  H04L65/1063 ,  H04L65/4076

Abstract: A method for a computer system includes receiving an identifier associated with a client streaming player, determining one or more attributes associated with the identifier from a data structure in response to the identifier, determining available channels from a plurality of channels that are to be made available to the client streaming player in response to the identifier, wherein available channels comprises a first channel, but not the second channel, determining a first graphical identifier associated with the first channel, providing the first graphical identifier to the client streaming player, receiving a user selection of the first graphical identifier indicating a user request to associate the first channel with the computer system, and providing an indicator of a server network address associated with the first channel to the client streaming player.

Abstract translation: 一种用于计算机系统的方法包括：接收与客户端流播放器相关联的标识符，响应于所述标识符从数据结构确定与所述标识符相关联的一个或多个属性，从多个信道中确定可用信道 响应于所述标识符到所述客户端流播放器，其中可用信道包括第一信道而不是所述第二信道，确定与所述第一信道相关联的第一图形标识符，向所述客户端流播放器提供所述第一图形标识符，接收用户 选择第一图形标识符，指示用户请求将第一频道与计算机系统相关联，以及向客户端流播放器提供与第一频道相关联的服务器网络地址的指示符。

公开(公告)号：US08706653B2

公开(公告)日：2014-04-22

申请号：US12963352

申请日：2010-12-08

Applicant: Gjergji Kasneci ,  Jurgen Anne Francois Marie Van Gael ,  Thore Kraepel ,  Ralf Herbrich ,  David Stern

Inventor： Gjergji Kasneci ,  Jurgen Anne Francois Marie Van Gael ,  Thore Kraepel ,  Ralf Herbrich ,  David Stern

IPC: G06F15/18

CPC classification number: G06N7/005

Abstract: Knowledge corroboration is described. In an embodiment many judges provide answers to many questions so that at least one answer is provided to each question and at least some of the questions have answers from more than one judge. In an example a probabilistic learning system takes features describing the judges or the questions or both and uses those features to learn an expertise of each judge. For example, the probabilistic learning system has a graphical assessment component which aggregates the answers in a manner which takes into account the learnt expertise in order to determine enhanced answers. In an example the enhanced answers are used for knowledge base clean-up or web-page classification and the learnt expertise is used to select judges for future questions. In an example the probabilistic learning system has a logical component that propagates answers according to logical relations between the questions.

Abstract translation: 描述知识佐证。 在一个实施例中，许多法官为许多问题提供答案，使得至少一个答案被提供给每个问题，并且至少一些问题具有来自多于一个法官的答案。 在一个例子中，概率学习系统采用描述法官或问题或两者的特征，并使用这些特征来学习每个法官的专业知识。 例如，概率学习系统具有图形评估组件，其以考虑到所学习的专业知识的方式聚集答案，以便确定增强的答案。 在一个例子中，增强的答案用于知识库清理或网页分类，并且学习的专业知识用于为将来的问题选择法官。 在一个例子中，概率学习系统具有根据问题之间的逻辑关系传播答案的逻辑组件。

公开(公告)号：US20120150771A1

公开(公告)日：2012-06-14

申请号：US12963352

申请日：2010-12-08

Applicant: Gjergji Kasneci ,  Jurgen Ann Francois Marie Van Gael ,  Thore Graepel ,  Ralf Herbrich ,  David Stern

Inventor： Gjergji Kasneci ,  Jurgen Ann Francois Marie Van Gael ,  Thore Graepel ,  Ralf Herbrich ,  David Stern

IPC: G06F15/18

CPC classification number: G06N7/005

Abstract: Knowledge corroboration is described. In an embodiment many judges provide answers to many questions so that at least one answer is provided to each question and at least some of the questions have answers from more than one judge. In an example a probabilistic learning system takes features describing the judges or the questions or both and uses those features to learn an expertise of each judge. For example, the probabilistic learning system has a graphical assessment component which aggregates the answers in a manner which takes into account the learnt expertise in order to determine enhanced answers. In an example the enhanced answers are used for knowledge base clean-up or web-page classification and the learnt expertise is used to select judges for future questions. In an example the probabilistic learning system has a logical component that propagates answers according to logical relations between the questions.

Abstract translation: 描述知识佐证。 在一个实施例中，许多法官为许多问题提供答案，使得至少一个答案被提供给每个问题，并且至少一些问题具有来自多于一个法官的答案。 在一个例子中，概率学习系统采用描述法官或问题或两者的特征，并使用这些特征来学习每个法官的专业知识。 例如，概率学习系统具有图形评估组件，其以考虑到所学习的专业知识的方式聚集答案，以便确定增强的答案。 在一个例子中，增强的答案用于知识库清理或网页分类，并且学习的专业知识用于为将来的问题选择法官。 在一个例子中，概率学习系统具有根据问题之间的逻辑关系传播答案的逻辑组件。

公开(公告)号：US20120101965A1

公开(公告)日：2012-04-26

申请号：US12912428

申请日：2010-10-26

Applicant: Philipp Hennig ,  David Stern ,  Thore Graepel ,  Ralf Herbrich

Inventor： Philipp Hennig ,  David Stern ,  Thore Graepel ,  Ralf Herbrich

IPC: G06F15/18 ,  G06N5/02

CPC classification number: G06N99/005 ,  G06N7/005

Abstract: Machine learning techniques may be used to train computing devices to understand a variety of documents (e.g., text files, web pages, articles, spreadsheets, etc.). Machine learning techniques may be used to address the issue that computing devices may lack the human intellect used to understand such documents, such as their semantic meaning. Accordingly, a topic model may be trained by sequentially processing documents and/or their features (e.g., document author, geographical location of author, creation date, social network information of author, and/or document metadata). Additionally, as provided herein, the topic model may be used to predict probabilities that words, features, documents, and/or document corpora, for example, are indicative of particular topics.

Abstract translation: 机器学习技术可用于训练计算设备以理解各种文档（例如，文本文件，网页，文章，电子表格等）。 可以使用机器学习技术来解决计算设备可能缺乏用于理解这样的文档的人类智力的问题，例如其语义意义。 因此，主题模型可以通过顺序处理文档和/或其特征（例如，文档作者，作者的地理位置，创作日期，作者的社交网络信息和/或文档元数据）来进行培训。 另外，如本文所提供的，主题模型可以用于预测词，特征，文档和/或文档语料库例如表示特定主题的概率。

公开(公告)号：US20110218946A1

公开(公告)日：2011-09-08

申请号：US12717109

申请日：2010-03-03

Applicant: David Stern ,  Ralf Herbrich ,  Milad Shokouhi ,  Thore Kurt Hartwig Graepel

Inventor： David Stern ,  Ralf Herbrich ,  Milad Shokouhi ,  Thore Kurt Hartwig Graepel

IPC: G06F17/30 ,  G06F15/18

CPC classification number: H04L12/1859 ,  G06F16/5866 ,  G06N20/00 ,  G06Q30/02 ,  G06Q50/01

Abstract: A user may request a presentation of a content item set, such as a social network comprising a set of status messages or an image database comprising a set of images. However, the volume and diversity of content items of the content item set may reduce the interest of the user in the presented content items. The potential interest of the user in the presented content items may be improved by selecting content items that are associated with one or more topics of potential interest to the user, and having a positive trending popularity among users of the content item set. Moreover, the interaction of the user with a presented content item may be monitored and used to determine the interest of the user in the topics associated with the presented content item and the popularity of the content item.

Abstract translation: 用户可以请求内容项目集的呈现，诸如包括一组状态消息的社交网络或包括一组图像的图像数据库。 然而，内容项目集的内容项目的数量和多样性可能降低用户在所呈现的内容项中的兴趣。 可以通过选择与用户潜在兴趣的一个或多个主题相关联的内容项，并且在内容项集合的用户中具有积极的趋势流行度来改善用户对所呈现的内容项目的潜在兴趣。 此外，可以监视用户与所呈现的内容项目的交互并用于确定用户在与所呈现的内容项目相关联的主题以及内容项目的受欢迎程度中的兴趣。

公开(公告)号：US20110131163A1

公开(公告)日：2011-06-02

申请号：US12628421

申请日：2009-12-01

Applicant: David Stern ,  Horst Cornelius Samulowitz ,  Ralf Herbrich ,  Thore Graepel

Inventor： David Stern ,  Horst Cornelius Samulowitz ,  Ralf Herbrich ,  Thore Graepel

IPC: G06F15/18 ,  G06N7/02

CPC classification number: G06N5/04 ,  G06Q10/00

Abstract: Managing a portfolio of experts is described where the experts may be for example, automated experts or human experts. In an embodiment a selection engine selects an expert from a portfolio of experts and assigns the expert to a specified task. For example, the selection engine has a Bayesian machine learning system which is iteratively updated each time an experts performance on a task is observed. For example, sparsely active binary task and expert feature vectors are input to the selection engine which maps those feature vectors to a multi-dimensional trait space using a mapping learnt by the machine learning system. In examples, an inner product of the mapped vectors gives an estimate of a probability distribution over expert performance. In an embodiment the experts are automated problem solvers and the task is a hard combinatorial problem such as a constraint satisfaction problem or combinatorial auction.

Abstract translation: 描述专家组合的描述，专家可能是例如，自动化专家或人类专家。 在一个实施例中，选择引擎从专家组合中选择专家，并将专家分配给指定的任务。 例如，选择引擎具有贝叶斯机器学习系统，每当观察到任务上的专家表现时，该学习系统被迭代地更新。 例如，将稀疏活动的二进制任务和专家特征向量输入到使用机器学习系统学习的映射将这些特征向量映射到多维特征空间的选择引擎。 在示例中，映射向量的内积给出了对专家性能的概率分布的估计。 在一个实施例中，专家是自动化问题解决者，并且任务是诸如约束满足问题或组合拍卖之类的硬组合问题。

公开(公告)号：US20100100416A1

公开(公告)日：2010-04-22

申请号：US12253854

申请日：2008-10-17

Applicant: Ralf Herbrich ,  Thore Graepel ,  David Stern

Inventor： Ralf Herbrich ,  Thore Graepel ,  David Stern

IPC: G06F17/30

CPC classification number: G06Q30/0633 ,  G06Q30/02 ,  G06Q30/0203

Abstract: A recommender system may be used to predict a user behavior that a user will give in relation to an item. In an embodiment such predictions are used to enable items to be recommended to users. For example, products may be recommended to customers, potential friends may be recommended to users of a social networking tool, organizations may be recommended to automated users or other items may be recommended to users. In an embodiment a memory stores a data structure specifying a bi-linear collaborative filtering model of user behaviors. In the embodiment an automated inference process may be applied to the data structure in order to predict a user behavior given information about a user and information about an item. For example, the user information comprises user features as well as a unique user identifier.

Abstract translation: 推荐系统可以用于预测用户将相对于项目给出的用户行为。 在一个实施例中，这样的预测用于使得可以向用户推荐项目。 例如，产品可能会推荐给客户，潜在的朋友可能会推荐给社交网络工具的用户，组织可能会推荐给自动化用户或其他项目可能推荐给用户。 在一个实施例中，存储器存储指定用户行为的双线性协同过滤模型的数据结构。 在该实施例中，自动推理过程可以应用于数据结构，以便预测给定关于用户的信息的用户行为和关于项目的信息。 例如，用户信息包括用户特征以及唯一的用户标识符。

公开(公告)号：US07647289B2

公开(公告)日：2010-01-12

申请号：US11421913

申请日：2006-06-02

Applicant: Thore K H Graepel ,  Ralf Herbrich ,  David Stern

Inventor： Thore K H Graepel ,  Ralf Herbrich ,  David Stern

IPC: G06F15/18 ,  G06F15/00

CPC classification number: A63F3/022 ,  A63F3/04 ,  G09B19/22

Abstract: We describe an apparatus for learning to predict moves in games such as chess, Go and the like, from historical game records. We obtain a probability distribution over legal moves in a given board configuration. This enables us to provide an automated game playing system, a training tool for players and a move selector/sorter for input to a game tree search system. We use a pattern extraction system to select patterns from historical game records. Our learning algorithm learns a distribution over the values of a move given a board position based on local pattern context. In another embodiment we use an Independent Bernoulli model whereby we assume each moved is played independently of other available moves.

Abstract translation: 我们描述一种从历史游戏记录中学习来预测诸如象棋，Go等游戏中的移动的装置。 在给定的电路板配置中，我们获得了合法移动的概率分布。 这使我们能够提供一种自动游戏系统，用于玩家的训练工具和用于输入到游戏树搜索系统的移动选择器/分选器。 我们使用模式提取系统从历史游戏记录中选择模式。 我们的学习算法基于局部模式上下文学习一个给定一个董事会职位的动作值的分布。 在另一个实施例中，我们使用独立的伯努利模型，由此我们假设每个移动都是独立于其他可用移动进行的。

公开(公告)号：US07628802B2

公开(公告)日：2009-12-08

申请号：US11333485

申请日：2006-01-17

Applicant: Jason White ,  David Stern

Inventor： Jason White ,  David Stern

IPC: A61F2/06

CPC classification number: A61F2/844 ,  A61F2/91 ,  A61F2/915 ,  A61F2002/91541 ,  A61F2002/91558 ,  A61F2002/91591 ,  A61F2220/0091 ,  Y10T403/32418

Abstract: A stent is constructed using interconnected links having micro-mechanical latching mechanisms. The micro-mechanical latching elements allow relative rotational movement of interconnected links in one rotational direction but restrict relative rotational movement of the two links in the opposite direction. The micro-mechanical latch surface features are formed using micro-electronic mechanical systems (MEMS) manufacturing methods. The male surface of the latching components contains an array of ridges or protrusions, and the receiving surface contains a matching array of recesses. The array of ridges or protrusions and the corresponding recesses have uniformly dissimilar slopes that result in a substantially greater frictional force in one direction than in the opposite direction. The separation distance between the two surfaces is such that the male latch surface is engaged with the receiving surface recesses in the low stress “locked” state, preventing motion in the undesired direction. Each male ridge or protrusion can be underlined by a void that promotes elastic deflection when sliding in the desired direction and recovery into the ‘locked ’ state when aligned with the recesses.

Abstract translation: 使用具有微机械闭锁机构的互连连接构造支架。 微机械闭锁元件允许互连连杆在一个旋转方向上的相对旋转运动，但限制两个连杆在相反方向上的相对旋转运动。 使用微电子机械系统（MEMS）制造方法形成微机械闩锁表面特征。 闩锁部件的阳表面包含脊或突起的阵列，并且接收表面包含匹配的凹槽阵列。 脊或突起的阵列和相应的凹槽具有均匀不相似的斜面，其在一个方向上产生比在相反方向上显着更大的摩擦力。 两个表面之间的分隔距离使得阳锁闩表面在低应力“锁定”状态下与接收表面凹部接合，从而防止在不期望的方向上的运动。 每个阳脊或突起可以由空隙加下划线，当空穴在所需方向上滑动时，该空隙促进弹性挠曲，并且当与凹槽对准时恢复到“锁定”状态。

公开(公告)号：US07618363B2

公开(公告)日：2009-11-17

申请号：US10636293

申请日：2003-08-06

Applicant: Jay Yadav ,  Mark Allen ,  David O'Brien ,  David Stern ,  Jason White ,  Michael A. Fonseca

Inventor： Jay Yadav ,  Mark Allen ,  David O'Brien ,  David Stern ,  Jason White ,  Michael A. Fonseca

IPC: A61M1/12

CPC classification number: A61M1/106 ,  A61M1/1008 ,  A61M1/1068 ,  A61M1/12

Abstract: A ventricular assist device comprises a sheet of hydraulically actuated material that can be affixed to prescribed locations on the surface of the heart to assist areas of the heart that do not contract normally. The material is comprised of a network of contractible unit cells that individually contract when fluid is pumped into them. These unit cells are connected together in a network that causes the sheet to contract radially inward. This contraction causes the sheet to transmit forces to the heart to assist in its natural contraction. A sensing function coordinates the contraction of the sheet with the contraction of the heart. The change in shape of the device is accomplished by distributing pressurized fluid throughout the spaces of the device by way of a network of channels. When pressure is removed from the fluid system, it assumes a deenergized “rest” position in which it does not transmit any forces to the surface of the heart. This property of the device prevents the device from inhibiting the heart's natural contractions in the event of a failure of the device or a loss of hydraulic power.

Abstract translation: 心室辅助装置包括一片液压致动材料片，该片材可以固定在心脏表面上的规定位置，以帮助心脏不能正常收缩的区域。 该材料由可收缩的单体电池网络组成，当将流体泵入其中时单独收缩。 这些单电池在网络中连接在一起，导致片材向内径向收缩。 这种收缩导致片材将力传递到心脏以帮助其自然收缩。 检测功能可以协调纸张与心脏收缩的收缩。 设备形状的变化通过通过通道网络将加压流体分配到设备的整个空间中来实现。 当压力从流体系统移除时，它承受断电的“静止”位置，在该位置，它不会将任何力传递到心脏表面。 该装置的这种特性防止装置在装置发生故障或液压动力损失的情况下抑制心脏的自然收缩。