-
公开(公告)号:US20080242420A1
公开(公告)日:2008-10-02
申请号:US11693635
申请日:2007-03-29
申请人: Thore Graepel , Ralf Herbrich , David Shaw
发明人: Thore Graepel , Ralf Herbrich , David Shaw
IPC分类号: A63F13/00
CPC分类号: A63F13/798 , A63F13/12 , A63F2300/5566 , A63F2300/63
摘要: Matchmaking processes at online game services often result in players having to wait unacceptably long times to receive a match or immediately receiving a poorly matched session. By using a matchmaking process which dynamically adapts a good balance is achieved between the quality of proposed matches (for example, in terms of how balanced, interesting and fun those matches are likely to be) and the waiting time for potential matches. A matchmaking threshold is specified. When a player seeks a match a waiting time is observed, for example, as to how long that player waits until starting a game or dropping out. Information about such waiting times is used to dynamically update the matchmaking threshold. The update is made on the basis of a relationship between information about the observed waiting time and a target waiting time. Further control may be achieved by using separate matchmaking thresholds and target waiting times for different game categories.
摘要翻译: 在线游戏服务中的对接过程通常会导致玩家不得不等待不可接受的长时间接收匹配或立即接收到不良匹配的会话。 通过使用匹配过程,动态地适应所提出的匹配的质量之间的良好平衡(例如,在这些匹配的平衡,有趣和有趣的可能性方面)以及潜在匹配的等待时间。 指定匹配阈值。 当玩家寻找比赛时,观察到等待时间,例如,玩家在开始游戏或退出之前等待多久。 关于这种等待时间的信息用于动态地更新配对阈值。 更新是根据观察到的等待时间的信息与目标等待时间之间的关系进行的。 可以通过对不同的游戏类别使用单独的配对阈值和目标等待时间来实现进一步的控制。
-
公开(公告)号:US20070192169A1
公开(公告)日:2007-08-16
申请号:US11693653
申请日:2007-03-29
申请人: Ralf Herbrich , Thore Graepel , David Shaw
发明人: Ralf Herbrich , Thore Graepel , David Shaw
CPC分类号: G06Q30/02 , G06Q30/0282
摘要: Reputation systems have been used to promote trust between participants in activities including online activities such as online market places. Existing online market places provide a reputation system which is a simple cumulative registry of user ratings on a given market place member. However, this simple system is open to abuse in situations where, for example, many positive ratings are given in a fraudulent manner. By modeling both reputation of participants and required reputation of participants an improved reputation system is provided. The required reputation may be thought of as a threshold, referred to herein as a required threshold, which may be used in determining how to update an indication of the reputation of a participant in the activity. The reputation system is able to learn information about required reputation and reputation of participants using an update process which is robust to participants who consistently give feedback of a particular type.
摘要翻译: 信誉系统已被用于促进参与者之间的信任,包括在线活动,如在线市场。 现有的在线市场地点提供了一个声誉系统,这是一个简单的累积登记的用户评级在给定的市场成员。 然而,这种简单的系统在例如以欺诈方式给出许多正面评级的情况下是可以滥用的。 通过对参与者的声誉和参与者的声誉进行建模,提供了改进的声誉系统。 所需的信誉可以被认为是阈值,这里称为所需阈值,其可用于确定如何更新活动中参与者的信誉的指示。 信誉系统能够使用对持续给出特定类型的反馈的参与者是稳健的更新过程来学习关于参与者的所需信誉和声誉的信息。
-
公开(公告)号:US07167849B2
公开(公告)日:2007-01-23
申请号:US10697727
申请日:2003-10-29
申请人: Thore Graepel , Ralf Herbrich
发明人: Thore Graepel , Ralf Herbrich
CPC分类号: G06K9/6255
摘要: An adaptive pattern classifier makes use of training patterns and a known non-linear invariance transformation to generate a classifier representation based on an infinite set of virtual training samples on a training trajectory. Given the non-linear invariance transformation, optimization can be formulated as a semidefinite program (SDP), which is given by a linear objective function that is minimized subject to a linear matrix inequality (LMI). In this manner, a small training set may be virtually supplemented using the non-linear invariance transformation to learn an effective classifier that satisfactorily recognizes patterns, even in the presence of known transformations that do not change the class of the pattern.
摘要翻译: 自适应模式分类器利用训练模式和已知的非线性不变性变换来生成基于训练轨迹上的虚拟训练样本的无限集合的分类器表示。 给定非线性不变性变换,可以将优化表达为半定程序(SDP),该方案由线性目标函数给出,该函数根据线性矩阵不等式(LMI)最小化。 以这种方式,可以使用非线性不变性变换来虚拟地补充小的训练集,以学习即使在不改变模式类别的已知变换的存在下令人满意地识别模式的有效分类器。
-
公开(公告)号:US20060184260A1
公开(公告)日:2006-08-17
申请号:US11276226
申请日:2006-02-17
申请人: Thore Graepel , Ralf Herbrich
发明人: Thore Graepel , Ralf Herbrich
IPC分类号: G06F19/00
CPC分类号: G06Q10/00 , G07F17/32 , G07F17/3232 , G07F17/3239 , G07F17/3276
摘要: Scoring of each player may be based on the outcomes of a game between players who compete against each other in one or more teams of one or more players. The scoring may also consider partial play where one or more players of a game only play for a portion of the full time of the game. Additionally or alternatively, the scoring may consider partial rankings of teams where the relative ranking of one team to another may be unknown. Each player's score may be represented as a distribution over potential scores which may indicate a confidence level in the distribution representing the player's score. The scoring may be used to track a player's progress and/or standing within the gaming environment, used in a leaderboard indication of rank, and/or may be used to match players with each other in a future game.
摘要翻译: 每个玩家的得分可以基于在一个或多个玩家的一个或多个队中彼此竞争的玩家之间的游戏的结果。 评分也可以考虑游戏的一个或多个玩家仅在游戏全部时间的一部分玩的部分游戏。 另外或者替代地,评分可以考虑对一队的相对排名与另一队的相对排名可能未知的队的部分排名。 每个玩家的得分可以表示为潜在分数的分布,其可以指示表示玩家得分的分布中的置信水平。 评分可以用于跟踪玩家在排行榜中使用的游戏环境中的进展和/或站立,并且/或可以用于在未来的游戏中将玩家彼此匹配。
-
公开(公告)号:US20050097068A1
公开(公告)日:2005-05-05
申请号:US10697727
申请日:2003-10-29
申请人: Thore Graepel , Ralf Herbrich
发明人: Thore Graepel , Ralf Herbrich
CPC分类号: G06K9/6255
摘要: An adaptive pattern classifier makes use of training patterns and a known non-linear invariance transformation to generate a classifier representation based on an infinite set of virtual training samples on a training trajectory. Given the non-linear invariance transformation, optimization can be formulated as a semidefinite program (SDP), which is given by a linear objective function that is minimized subject to a linear matrix inequality (LMI). In this manner, a small training set may be virtually supplemented using the non-linear invariance transformation to learn an effective classifier that satisfactorily recognizes patterns, even in the presence of known transformations that do not change the class of the pattern.
摘要翻译: 自适应模式分类器利用训练模式和已知的非线性不变性变换来生成基于训练轨迹上的虚拟训练样本的无限集合的分类器表示。 给定非线性不变性变换,可以将优化表达为半定程序(SDP),该方案由线性目标函数给出,该函数根据线性矩阵不等式(LMI)最小化。 以这种方式,可以使用非线性不变性变换来虚拟地补充小的训练集,以学习即使在不改变模式类别的已知变换的存在下令人满意地识别模式的有效分类器。
-
公开(公告)号:US08672764B2
公开(公告)日:2014-03-18
申请号:US11693635
申请日:2007-03-29
申请人: Thore Graepel , Ralf Herbrich , David Shaw
发明人: Thore Graepel , Ralf Herbrich , David Shaw
IPC分类号: A63F9/24
CPC分类号: A63F13/798 , A63F13/12 , A63F2300/5566 , A63F2300/63
摘要: Matchmaking processes at online game services often result in players having to wait unacceptably long times to receive a match or immediately receiving a poorly matched session. By using a matchmaking process which dynamically adapts a good balance is achieved between the quality of proposed matches (for example, in terms of how balanced, interesting and fun those matches are likely to be) and the waiting time for potential matches. A matchmaking threshold is specified. When a player seeks a match a waiting time is observed, for example, as to how long that player waits until starting a game or dropping out. Information about such waiting times is used to dynamically update the matchmaking threshold. The update is made on the basis of a relationship between information about the observed waiting time and a target waiting time. Further control may be achieved by using separate matchmaking thresholds and target waiting times for different game categories.
摘要翻译: 在线游戏服务中的对接过程通常会导致玩家不得不等待不可接受的长时间接收匹配或立即接收到不良匹配的会话。 通过使用匹配过程,动态地适应所提出的匹配的质量之间的良好平衡(例如,在这些匹配的平衡,有趣和有趣的可能性方面)以及潜在匹配的等待时间。 指定匹配阈值。 当玩家寻找比赛时,观察到等待时间,例如,玩家在开始游戏或退出之前等待多久。 关于这种等待时间的信息用于动态地更新配对阈值。 更新是根据观察到的等待时间的信息与目标等待时间之间的关系进行的。 可以通过对不同的游戏类别使用单独的配对阈值和目标等待时间来实现进一步的控制。
-
公开(公告)号:US08645298B2
公开(公告)日:2014-02-04
申请号:US12912428
申请日:2010-10-26
申请人: Philipp Hennig , David Stern , Thore Graepel , Ralf Herbrich
发明人: Philipp Hennig , David Stern , Thore Graepel , Ralf Herbrich
CPC分类号: G06N99/005 , G06N7/005
摘要: 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.
摘要翻译: 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:US08433660B2
公开(公告)日:2013-04-30
申请号:US12628421
申请日:2009-12-01
摘要: 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.
摘要翻译: 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:US20120089446A1
公开(公告)日:2012-04-12
申请号:US12899566
申请日:2010-10-07
申请人: Anoop Gupta , Thore Graepel , Ralf Herbrich
发明人: Anoop Gupta , Thore Graepel , Ralf Herbrich
IPC分类号: G06Q30/00
CPC分类号: G06Q30/0208 , G06Q30/0601 , G06Q50/01
摘要: A publishing engine captures commercial information associated with a first user and automatically notifies other users in the first user's social network of this commercial information. The first user authorizes an e-commerce system to access his or her social network and to publish commercial information about the first user's commercial activity (e.g., a purchase or other commercial transaction) to users in the social network. By this automated notification, the notified users in the first user's social network can learn that the first user has completed a commercial transaction pertaining to a particular product or service. If a notified user is interested in a similar product or service, he or she can contact the first user to inquire about the first user's experience and information with the product or service.
摘要翻译: 发布引擎捕获与第一用户相关联的商业信息,并自动通知第一用户的社交网络中的其他用户该商业信息。 第一个用户授权电子商务系统访问他或她的社交网络,并且向社交网络中的用户发布关于第一用户的商业活动(例如,购买或其他商业交易)的商业信息。 通过该自动通知,第一用户社交网络中通知的用户可以知道第一用户已经完成了与特定产品或服务有关的商业交易。 如果通知用户对类似的产品或服务感兴趣,他或她可以联系第一个用户以查询第一个用户的体验和产品或服务的信息。
-
公开(公告)号:US20090093287A1
公开(公告)日:2009-04-09
申请号:US11869165
申请日:2007-10-09
申请人: Ralf Herbrich , Thore Graepel , Thomas Minka , Pierre Dangauthier
发明人: Ralf Herbrich , Thore Graepel , Thomas Minka , Pierre Dangauthier
IPC分类号: A63F9/24
CPC分类号: G06N7/005
摘要: A process for determining relative player skills and draw margins is described. Information about an outcome of a game between at least a first player opposing a second player is received. Also, for each player, skill statistics are received associated with a distribution representing belief about skill of that player. Draw margin statistics are received associated with a distribution representing belief about ability of that player to force a draw. An update process is performed to update the statistics on the basis of the received information about the game outcome. In an embodiment a Bayesian inference process is used during the update process which may take past and future player achievement into account.
摘要翻译: 描述了确定相对玩家技能和绘制边距的过程。 接收关于至少与第二玩家对抗的第一玩家之间的游戏结果的信息。 此外,对于每个玩家,接收与表示对该玩家的技能的信念的分布相关联的技能统计。 收到的边际统计数据收到与表示对该玩家强制抽签能力相关的分布相关联。 执行更新处理以基于所接收的关于游戏结果的信息更新统计信息。 在一个实施例中,在更新过程中使用贝叶斯推理过程,其可以考虑过去和未来的玩家成就。
-
-
-
-
-
-
-
-
-