公开(公告)号：US12124965B2

公开(公告)日：2024-10-22

申请号：US17114408

申请日：2020-12-07

Applicant: International Business Machines Corporation

Inventor： Hamid Dadkhahi ,  Karthikeyan Shanmugam ,  Jesus Maria Rios Aliaga ,  Payel Das ,  Samuel Chung Hoffman

IPC: G06N5/01 ,  G06N3/126

CPC classification number: G06N5/01 ,  G06N3/126

Abstract: Aspects of the present invention disclose a method, computer program product, and system for optimizing a result for a combinatorial optimization problem. The method includes one or more processors receiving a black-box model. The method further includes one or more processors learning a multilinear polynomial surrogate model employing an exponential weight update rule. The method further includes one or more processors optimizing the learnt multilinear polynomial surrogate model. The method further includes one or more processors applying the black-box model to the optimized solution found by the multilinear polynomial surrogate model. In an additional aspect, the method of learning an optimized multilinear polynomial surrogate model employing an exponential weight update rule further includes one or more processors calculating utilizing data from the black-box model, an update of the coefficients of the multilinear polynomial surrogate model.

公开(公告)号：US12105243B2

公开(公告)日：2024-10-01

申请号：US17723920

申请日：2022-04-19

Applicant: Halliburton Energy Services, Inc.

Inventor： Dingding Chen ,  Mark A. Proett ,  Li Gao ,  Christopher Michael Jones

IPC: G01V20/00 ,  E21B49/08 ,  G06N3/126 ,  G06N20/00

CPC classification number: G01V20/00 ,  E21B49/087 ,  G06N3/126 ,  G06N20/00

Abstract: Improved systematic inversion methodology applied to formation testing data interpretation with spherical, radial and/or cylindrical flow models is disclosed. A method of determining a flow line parameter includes determining a diverse set of flow models and selecting at least one flow model from the diverse set of flow models representative, at least in part, of a formation tester tool, at least one formation, at least one fluid, and at least one flow of the at least one fluid. The method further includes lowering the formation testing tool into the at least one formation to intersect with the formation at least one formation and sealing a probe of the formation tester placed in fluid communication with the at least one formation. The method further includes initiating flow from the at least one formation and utilizing the at least one selected flow model to predict the flow line parameter.

公开(公告)号：US12099911B2

公开(公告)日：2024-09-24

申请号：US18076494

申请日：2022-12-07

Applicant: Strong Force IoT Portfolio 2016, LLC

Inventor： Charles Howard Cella ,  Gerald William Duffy, Jr. ,  Jeffrey P. McGuckin ,  Mehul Desai

IPC: G06N3/006 ,  B62D15/02 ,  G01M13/028 ,  G01M13/04 ,  G01M13/045 ,  G05B13/02 ,  G05B19/418 ,  G05B23/02 ,  G06F18/21 ,  G06N3/02 ,  G06N3/044 ,  G06N3/045 ,  G06N3/047 ,  G06N3/084 ,  G06N3/088 ,  G06N5/046 ,  G06N7/01 ,  G06N20/00 ,  G06Q10/04 ,  G06Q10/0639 ,  G06Q30/02 ,  G06Q30/06 ,  G06Q50/00 ,  G06V10/778 ,  G06V10/82 ,  G16Z99/00 ,  H02M1/12 ,  H03M1/12 ,  H04B17/23 ,  H04B17/309 ,  H04B17/318 ,  H04B17/345 ,  H04L1/00 ,  H04L1/18 ,  H04L1/1867 ,  H04L67/1097 ,  H04L67/12 ,  H04W4/38 ,  H04W4/70 ,  B62D5/04 ,  G05B19/042 ,  G06F17/18 ,  G06F18/25 ,  G06N3/126 ,  H01B17/40 ,  H04B17/29 ,  H04B17/40 ,  H04L5/00 ,  H04L67/306

CPC classification number: G06N3/006 ,  B62D15/0215 ,  G01M13/028 ,  G01M13/04 ,  G01M13/045 ,  G05B13/028 ,  G05B19/4183 ,  G05B19/4184 ,  G05B19/41845 ,  G05B19/4185 ,  G05B19/41865 ,  G05B19/41875 ,  G05B23/0221 ,  G05B23/0229 ,  G05B23/024 ,  G05B23/0264 ,  G05B23/0283 ,  G05B23/0286 ,  G05B23/0289 ,  G05B23/0291 ,  G05B23/0294 ,  G05B23/0297 ,  G06F18/2178 ,  G06N3/02 ,  G06N3/044 ,  G06N3/045 ,  G06N3/047 ,  G06N3/084 ,  G06N3/088 ,  G06N5/046 ,  G06N7/01 ,  G06N20/00 ,  G06Q10/04 ,  G06Q10/0639 ,  G06Q30/02 ,  G06Q30/0278 ,  G06Q30/06 ,  G06Q50/00 ,  G06V10/7784 ,  G06V10/82 ,  G16Z99/00 ,  H02M1/12 ,  H03M1/12 ,  H04B17/23 ,  H04B17/309 ,  H04B17/318 ,  H04B17/345 ,  H04L1/0002 ,  H04L1/0041 ,  H04L1/18 ,  H04L1/1874 ,  H04L67/1097 ,  H04L67/12 ,  H04W4/38 ,  H04W4/70 ,  B62D5/0463 ,  G05B19/042 ,  G05B23/02 ,  G05B23/0208 ,  G05B2219/32287 ,  G05B2219/35001 ,  G05B2219/37337 ,  G05B2219/37351 ,  G05B2219/37434 ,  G05B2219/37537 ,  G05B2219/40115 ,  G05B2219/45004 ,  G05B2219/45129 ,  G06F17/18 ,  G06F18/21 ,  G06F18/217 ,  G06F18/25 ,  G06N3/126 ,  H01B17/40 ,  H04B17/29 ,  H04B17/40 ,  H04L1/0009 ,  H04L5/0064 ,  H04L67/306 ,  Y02P80/10 ,  Y02P90/02 ,  Y02P90/80 ,  Y04S50/00 ,  Y04S50/12 ,  Y10S707/99939

Abstract: System and methods for learning data patterns predictive of an outcome are described. An example system may include a plurality of input sensors communicatively coupled to a controller; a data collection circuit structured to collect output data from the plurality of input sensors; and a machine learning data analysis circuit structured to receive the output data, learn received output data patterns indicative of an outcome, and learn a preferred input data collection band among a plurality of available input data collection bands. The machine learning data analysis circuit may be structured to learn received output data patterns by being seeded with a model based on industry-specific feedback. The outcome may be at least one of: a reaction rate, a production volume, or a required maintenance.

公开(公告)号：US12099152B2

公开(公告)日：2024-09-24

申请号：US17704520

申请日：2022-03-25

Applicant: Ge Wang ,  Mengzhou Li ,  David S. Rundle

Inventor： Ge Wang ,  Mengzhou Li ,  David S. Rundle

IPC: G01T1/17 ,  G01T1/36 ,  G06N3/04 ,  G06N3/045 ,  G06N3/047 ,  G06N3/08 ,  G06N3/088 ,  G06N3/126

CPC classification number: G01T1/17 ,  G01T1/171 ,  G01T1/36 ,  G06N3/045 ,  G06N3/047 ,  G06N3/08 ,  G06N3/088 ,  G06N3/126

Abstract: A method for x-ray photon-counting data correction. The method includes generating, by a training data generation module, training input spectral projection data based, at least in part, on a reference spectral projection data. The training input spectral projection data includes at least one of a pulse pileup distortion, a charge splitting distortion, and/or noise. The method further includes training, by a training module, a data correction artificial neural network (ANN) based, at least in part, on training data. The data correction ANN includes a pulse pileup correction ANN, and a charge splitting correction ANN. The training data includes the training input spectral projection data and the reference spectral projection data.

公开(公告)号：US12001963B2

公开(公告)日：2024-06-04

申请号：US17207340

申请日：2021-03-19

Applicant: Sensormatic Electronics, LLC

Inventor： Gopi Subramanian ,  Michael C. Stewart

IPC: G06N3/126 ,  G06F18/214 ,  G06N20/00 ,  G08B13/24

CPC classification number: G06N3/126 ,  G06F18/214 ,  G06N20/00 ,  G08B13/2474

Abstract: Example aspects include techniques for building a ML model in a use case with prohibitive training data and employing the ML model within the use case. These techniques may include determining training information including a plurality of stray training reads and a plurality of valid training reads, determining modified training information based at least in part on modifying the plurality of valid training reads, and generating a model for distinguishing a valid read from a stray read based on the modified training information and an evolutionary algorithm. In addition, the techniques may include detecting, by a monitoring device, a plurality of tag reads in response to a plurality of interactions between a tag and the monitoring device, and determining, by the monitoring device, a plurality of valid tag reads based on the model and plurality of tag reads.

公开(公告)号：US11989655B2

公开(公告)日：2024-05-21

申请号：US17079677

申请日：2020-10-26

Applicant: JIANGXI UNIVERSITY OF SCIENCE AND TECHNOLOGY

Inventor： Xiaoyan Luo ,  Hui Yu ,  Tao Deng ,  Junxi Liu ,  Xuetao Zhang

IPC: G06N3/084 ,  G06N3/04 ,  G06N3/082 ,  G06N3/126 ,  G06N5/022 ,  G10L25/03 ,  G10L25/30 ,  G10L25/51

CPC classification number: G06N3/084 ,  G06N3/04 ,  G06N3/082 ,  G06N3/126 ,  G06N5/022 ,  G10L25/03 ,  G10L25/30 ,  G10L25/51

Abstract: Embodiments of the present application provide a prediction method, device and system for rock mass instability stages, and belong to the technical field of rock mass instability prediction. The method includes the steps: acquiring acoustic emission signals of rock mass; extracting feature parameters from the acquired acoustic emission signals; and predicting instability stages of the rock mass in accordance with the feature parameters and a preset back propagation (BP) neural network model, wherein the preset BP neural network model is obtained by training a BP neural network and a genetic algorithm by virtue of the feature parameters of the acoustic emission signals at different rock mass instability stages. According to the technical solution in the present application, the problem in the training process of the BP neural network model that model parameter optimization may be easily trapped in a locally optimal solution is effectively solved.

公开(公告)号：US20240070472A1

公开(公告)日：2024-02-29

申请号：US17932290

申请日：2022-09-14

Applicant: Inventec (Pudong) Technology Corporation ,  INVENTEC CORPORATION

Inventor： Ying-Sheng LUO ,  Trista Pei-Chun CHEN ,  Li-Ya SU ,  Ching Hui LI

IPC: G06N3/12 ,  B65B57/00

CPC classification number: G06N3/126 ,  B65B57/00

Abstract: The present disclosure provides a packing method including following steps. Genetic algorithm is utilized to calculate multiple packing programs. Multiple candidate packing programs including all items are selected from the packing programs. Among each of the candidate packing programs, at least one of the items to be placed earlier is classified into a first subset, and at least another one of the items to be placed later is classified into a second subset. Among each of the candidate packing programs, a first packing for the first subset is maintained, and a second packing for the second subset is recalculated by using a greedy algorithm to generate an updated second packing.

公开(公告)号：US20240068907A1

公开(公告)日：2024-02-29

申请号：US18021493

申请日：2022-05-11

Applicant: DALIAN UNIVERSITY OF TECHNOLOGY

Inventor： Ximing SUN ,  Aina WANG ,  Yingshun LI ,  Pan QIN ,  Chongquan ZHONG

IPC: G01M13/045 ,  G06N3/126

CPC classification number: G01M13/045 ,  G06N3/126

Abstract: The present invention provides an optimization algorithm for automatically determining variational mode decomposition parameters based on bearing vibration signals. First, mode energy is used to reflect bandwidth, a bandwidth optimization sub-model is established to automatically obtain optimal bandwidth parameter αopt. Secondly, energy loss optimization sub-model is established to avoid under-decomposition. Thirdly, a mode mean position distance optimization sub-model is established to prevent the generation of too much K and avoid the phenomenon of over-decomposition. Finally, considering the interaction between the bandwidth parameter α and the total number of modes K, the interaction between mode components and the integrity of reconstruction information, nonlinear transformation is performed by a logarithmic function, so as to make the values of three optimization sub-models form similar scales, obtain an optimization model that can automatically determine optimal VMD parameters αopt and Kopt, and establish a quantitative evaluation index for the decomposition performance of a VMD algorithm.

公开(公告)号：US11915798B2

公开(公告)日：2024-02-27

申请号：US16886307

申请日：2020-05-28

Applicant: FUJITSU LIMITED

Inventor： Hideyuki Jippo

IPC: G16C20/30 ,  G06N3/12 ,  G06N3/126 ,  G06N5/022 ,  G06N5/04 ,  G06N5/048 ,  G06N20/00 ,  G06N20/10 ,  G16C20/40 ,  G16C20/70

CPC classification number: G16C20/30 ,  G06N3/126 ,  G06N5/048 ,  G16C20/70 ,  G06N3/12 ,  G06N5/022 ,  G06N5/04 ,  G06N20/00 ,  G06N20/10 ,  G16C20/40

Abstract: An apparatus includes a memory and a processor coupled to the memory. The processor is configured to: determine a degree of similarity between a target material and a first material based on a structure and characteristic of each of the target material and the first material; predict a characteristic value of the target material based on a first value representing the characteristic of the first material; and output the predicted characteristic value.

公开(公告)号：US20240054368A1

公开(公告)日：2024-02-15

申请号：US17818472

申请日：2022-08-09

Applicant: WORKDAY, INC.

Inventor： Volodymyr TOMENKO ,  Dalmo CIRNE ,  Ganesh RAJARATNAM ,  Chris CHEN

IPC: G06N5/04 ,  G06N3/12

CPC classification number: G06N5/041 ,  G06N3/126

Abstract: In some aspects, the techniques described herein relate to a method including: initializing a population of hypotheses; computing misfit values for each of the hypotheses, the misfit values computed using a fitness function including a weighted summation, wherein terms of weighted summation include metric functions; generating a plurality of offspring hypotheses based on the population of hypotheses and a crossover bitmask; generating a new population using the plurality of offspring and a subset of the population of hypotheses; mutating at least one hypothesis in the new population; selecting a hypothesis from the new population based on a corresponding misfit value of the hypothesis; and allocating at least one resource based on the hypothesis.