公开(公告)号：US20140343903A1

公开(公告)日：2014-11-20

申请号：US13898087

申请日：2013-05-20

Applicant: NEC CORPORATION

Inventor： Ryohei FUJIMAKI ,  Shaohua LI

IPC: G06F17/16

CPC classification number: G06F17/18 ,  G06K9/6297 ,  G06N7/005

Abstract: An approximate computation unit computes an approximate of a determinant of a Hessian matrix relating to a parameter of an observation model represented as a linear combination of parameters determined by each layer 1 latent variable of factorial hidden Markov models. A variational probability computation unit computes a variational probability of a latent variable using the approximate of the determinant. A latent state removal unit removes a latent state based on a variational distribution. A parameter optimization unit optimizes the parameter for a criterion value that is defined as a lower bound of an approximate obtained by Laplace-approximating a marginal log-likelihood function with respect to an estimator for a complete variable, and computes the criterion value. A convergence determination unit determines whether or not the criterion value has converged.

Abstract translation: 近似计算单元计算与由表示为由阶乘隐马尔可夫模型的每个层1潜在变量确定的参数的线性组合的观测模型的参数相关的Hessian矩阵的行列式的近似值。 变分概率计算单元使用行列式的近似来计算潜在变量的变分概率。 潜在去除单元基于变分分布去除潜在状态。 一个参数优化单元优化一个标准值的参数，该标准值被定义为通过拉普拉斯逼近一个关于完整变量的估计器的边际对数似然函数获得的近似值的下限，并计算标准值。 收敛确定单元确定标准值是否收敛。

公开(公告)号：US20130326532A1

公开(公告)日：2013-12-05

申请号：US13653464

申请日：2012-10-17

Applicant: NEC CORPORATION

Inventor： Ryohei FUJIMAKI ,  Kouhei HAYASHI

IPC: G06F9/50

CPC classification number: G06F9/50

Abstract: A parallel allocation calculating unit calculates a parallel allocation candidate which is an element candidate in target data allocated per processing performed in parallel. A parallel calculation amount estimation processing unit estimates the calculation amount required for parallel processing when a parallel allocation candidate is allocated, based on a nonzero element count in the target data. An optimality decision processing unit decides whether or not the parallel allocation candidate is optimal based on the calculated calculation amount, and allocates the optimal element per processing performed in parallel.

Abstract translation: 并行分配计算单元计算作为并行执行的每个处理分配的目标数据中的元素候选的并行分配候选。 并行计算量估计处理单元基于目标数据中的非零元素计数来估计分配并行分配候选的并行处理所需的计算量。 最优性判定处理单元基于计算出的计算量判定并行分配候选是否为最优，并且分配并行执行的每个处理的最优元素。

公开(公告)号：US20210103472A1

公开(公告)日：2021-04-08

申请号：US16608113

申请日：2018-04-09

Applicant: NEC CORPORATION

Inventor： Akihiro YABE ,  Masato ASAHARA ,  Ryohei FUJIMAKI

IPC: G06F9/50 ,  G06F9/48

Abstract: A computational resource management apparatus is for managing a cluster system that executes a plurality of tasks. The computational resource management apparatus includes a condition specification unit that specifies a relationship between computational resources of the cluster system and computation time, a dependency relationship between tasks, and an execution time limit of each task, and a scheduling unit that determines, for each task, an execution sequence and computational resources to be allocated from among the computational resources of the cluster system, based on the relationship between the computational resources and computation time and the dependency relationship that are specified, such that the execution time limit is met.

公开(公告)号：US20200057948A1

公开(公告)日：2020-02-20

申请号：US16346004

申请日：2017-10-05

Applicant: NEC CORPORATION

Inventor： Ryohei FUJIMAKI ,  Yukitaka KUSUMURA ,  Masato ASAHARA ,  Yusuke MURAOKA

IPC: G06N5/04 ,  G06F16/22 ,  G06N20/00

Abstract: A feature design unit 81 designs, from relational data, a feature as a variable likely to affect an objective variable. A feature generating unit 82 generates the designed feature, from the relational data. A learning unit 83 learns a prediction model, on the basis of the generated feature.

公开(公告)号：US20190079796A1

公开(公告)日：2019-03-14

申请号：US16085097

申请日：2017-04-27

Applicant: NEC Corporation

Inventor： Masato ASAHARA ,  Akihiro YABE ,  Kyota KANNO ,  Ryohei FUJIMAKI

IPC: G06F9/48 ,  G06F9/50 ,  G06F11/34 ,  G06N7/00

Abstract: Computational resource management device includes a model learning unit that uses a measured value of an execution time of data processing, a measured value of a deresource amount, and a feature of input data as training data to learn a model indicating relationship between the execution time and the resource, an execution time estimation unit that inputs, into the model, a feature of data scheduled to be input to calculate an estimated value of the execution time of the scheduled data processing, a resource amount calculation unit that uses the estimated value, a variation index indicating variation in the estimated value, and distribution of estimated residuals to calculate resource amount required in the scheduled data processing, and an execution plan creation unit that creates an execution plan of the scheduled data processing, based on the estimated value, the variation index, the distribution of estimated residuals, and the calculated resource amount.

公开(公告)号：US20180285787A1

公开(公告)日：2018-10-04

申请号：US15764681

申请日：2016-08-09

Applicant: NEC CORPORATION

Inventor： Shinji ITO ,  Ryohei FUJIMAKI

IPC: G06Q10/04 ,  G06F17/18 ,  G06F15/18 ,  G06F17/11

Abstract: A model input unit 84 receives a linear regression model represented by a function having an objective variable as an explanatory variable. A candidate point input unit 85 receives, for the objective variable included in the linear regression model, at least one candidate point which is a discrete candidate for a possible value of the objective variable. An optimization unit 86 calculates the objective variable that optimizes an objective function having the linear regression model as an argument.

公开(公告)号：US20180268425A1

公开(公告)日：2018-09-20

申请号：US15756052

申请日：2016-08-09

Applicant: NEC CORPORATION

Inventor： Yusuke MURAOKA ,  Ryohei FUJIMAKI

IPC: G06Q30/02 ,  G06F17/18

CPC classification number: G06Q30/0201 ,  G06F17/18 ,  G06Q30/0202

Abstract: An information processing apparatus 100 includes: a reception unit 10 that receives analysis subjects and variables relating to the analysis subjects; and a graph generation unit 20 that specifies degrees of influence that the variables have on the analysis subjects with the degrees of influence divided into positive and negative, and generates a graph indicating the specified degrees of positive influence and the specified degrees of negative influence as distances between the variables and the analysis subjects.

公开(公告)号：US20180267934A1

公开(公告)日：2018-09-20

申请号：US15755214

申请日：2016-09-07

Applicant: NEC Corporation

Inventor： Ryohei FUJIMAKI ,  Yusuke MURAOKA

IPC: G06F17/11 ,  G06F17/16 ,  G06Q10/04

CPC classification number: G06F17/11 ,  G06F17/16 ,  G06Q10/04 ,  G06Q10/063

Abstract: An information processing device according to the present invention includes: a problem generator that, based on a first optimization problem, a lower-dimensional expression that is an expression for approximating uncertain data for the first optimization problem at a lower dimension than a dimension of the uncertain data, and a first data region that is a region of the uncertain data, generates a second optimization problem into that the first optimization problem is transformed in such a way that the second optimization problem relates to the lower-dimensional expression, and a second data region into that the first data region is transformed; and a problem solver that computes an optimum solution to the second optimization problem by using the second data region.

公开(公告)号：US20180025072A1

公开(公告)日：2018-01-25

申请号：US15549897

申请日：2016-01-27

Applicant: NEC Corporation

Inventor： Ryohei FUJIMAKI ,  Yousuke MOTOHASHI

IPC: G06F17/30 ,  G06Q30/02 ,  G06N7/00

CPC classification number: G06F16/285 ,  G06F16/2246 ,  G06N5/003 ,  G06N7/005 ,  G06N20/00 ,  G06Q30/0201

Abstract: A classifier 81 classifies target data into a cluster on the basis of a mixture model defined using two different types of variables that indicate features of the target data. In this classification, the classifier 81 classifies the target data into a cluster on the basis of a mixture model in which a mixing ratio of the mixture model is represented by a function of a first variable and in which the element distribution of the clusters into which the target data is classified is represented by a function of a second variable.

公开(公告)号：US20170076307A1

公开(公告)日：2017-03-16

申请号：US15125267

申请日：2015-02-27

Applicant: NEC Corporation

Inventor： Yosuke MOTOHASHI ,  Satoshi MORINAGA ,  Ryohei FUJIMAKI ,  Riki ETO ,  Masato ASAHARA

IPC: G06Q30/02

CPC classification number: G06Q30/0206 ,  G06N7/005 ,  G06Q30/0202

Abstract: A price estimation device that can predict a price with a high degree of precision is disclosed. Said price estimation device has a price-predicting means that predicts a price pertaining to second information in a target second time period by applying rule information to said second information, which includes explanatory variables. Said rule information represents the relationship between the explanatory variables and the price, said relationship having been extracted on the basis of a first-information set comprising first information in which explanatory-variable values are associated with price values. The explanatory variables include an attribute that represents a length of time, determined on the basis of a first time period in which a specific event occurs, pertaining to a target object associated with the aforementioned first information or the abovementioned second information. The value of said attribute in the second information is the length of time between the first time period and the second time period, and the value of the attribute in the first information is the length of time between the first time period and a third time period associated with the abovementioned price.

Abstract translation: 公开了可以高精度地预测价格的价格估计装置。 所述价格估计装置具有价格预测装置，其通过将规则信息应用于包括解释变量的所述第二信息来预测与目标第二时间段中的第二信息有关的价格。 所述规则信息表示解释变量和价格之间的关系，所述关系是基于包括解释变量值与价格值相关联的第一信息的第一信息集来提取的。 解释性变量包括表示基于与特定事件发生的第一时间段相关的与上述第一信息相关联的目标对象或上述第二信息确定的时间长度的属性。 第二信息中的所述属性的值是第一时间段和第二时间段之间的时间长度，第一信息中的属性值是第一时间段与第三时间段之间的时间长度 与上述价格相关。