Machine Learning Patents (Class 706/12)
  • Patent number: 10742516
    Abstract: Systems, methods, and computer-readable media for distributing machine learning. In some examples, a first GAN model is deployed to a first network edge device and a second GAN model is deployed to a second network edge device. A generator of the first GAN model can be trained using real telemetry data of a first computing node and a generator of the second GAN model can be trained using real telemetry data of a second IoT device. The generator of the first GAN model and the generator of the second GAN model can be received. Additionally, a unified generator of a unified GAN model can be trained using the generator of the first GAN model and the generator of the second GAN model. Subsequently, the unified GAN model can be deployed to a third computing node for monitoring operation of the third IoT device.
    Type: Grant
    Filed: February 6, 2019
    Date of Patent: August 11, 2020
    Assignee: CISCO TECHNOLOGY, INC.
    Inventors: Volodymyr Iashyn, Borys Viacheslavovych Berlog, Dmitri Goloubev
  • Patent number: 10740378
    Abstract: An information processing device according to an embodiment includes one or more processors. The processors perform hierarchical clustering of a key phrase group. The processors divide the key phrase group into candidate clusters. The processors receive a selectin operation of one item from predetermined items for classifying the document group. The processors calculate, for each candidate cluster, a score indicating utility with respect to the selected item. The processors decide, as a reference cluster, a candidate cluster for which the score has a predetermined ranking. The processors divide the reference cluster into sub-clusters. The processors extract predetermined sub-items in the lower levels of the selected item. And the processors control presentation of an expansion image for expressing the information volume of the documents for each sub-item and each sub-cluster.
    Type: Grant
    Filed: February 20, 2018
    Date of Patent: August 11, 2020
    Assignee: Kabushiki Kaisha Toshiba
    Inventors: Kosei Fume, Tomohiro Yamasaki
  • Patent number: 10740619
    Abstract: A media item comprising a set of frames is received by a feature extraction system. A frame predictor is executed on each frame of the set of frames. An error representation is extracted for each frame of the set of frames during the execution of the frame predictor. An error-based feature vector is generated from the error representations associated with each frame of the set of frames. A seed media item is identified having a first error-based feature vector. A similarity score is determined among the first error-based feature vector and each error-based feature vector of a set of error-based feature vectors. A subset of error-based feature vectors, hence a subset of corresponding media items, is selected based on similarity score.
    Type: Grant
    Filed: August 31, 2018
    Date of Patent: August 11, 2020
    Assignee: Uber Technologies, Inc.
    Inventors: Pegah Massoudifar, Vijay Mahadevan, Jonathan Berliner
  • Patent number: 10740568
    Abstract: An example embodiment may involve obtaining an incident record relating to a user. The embodiment may also involve generating and providing, for display on a graphical user interface, a single window including a dialog region, an incident record region, and a suggestion region. The embodiment may also involve determining candidate messages by incorporating components of the incident record into predetermined message templates. The embodiment may also involve determining a scoring for the candidate messages based on a relevance to a conversation between the user and an agent. The embodiment may also involve based on the scoring, selecting one or more of the candidate messages to include in a set of suggested messages displayed in the suggestion region. The embodiment may also involve receiving input from the agent selecting one of the suggested messages, and then responsively displaying the selected suggested message as part of the conversation in the dialog region.
    Type: Grant
    Filed: January 24, 2018
    Date of Patent: August 11, 2020
    Assignee: ServiceNow, Inc.
    Inventor: Troy Azmoon
  • Patent number: 10742642
    Abstract: In some examples, a system for authenticating users can include a processor to train a first predictive application based on a first set of user engagements with advertisements, wherein the first predictive application is associated with a first advertising identifier. The processor can also train a second predictive application based on a second set of user engagements with the advertisements, wherein the second predictive application is associated with a second advertising identifier. Additionally, the processor can compare the first predictive application and the second predictive application and authenticate a user in response to detecting a similarity of the first predictive application and the second predictive application is below a threshold value, wherein authenticating the user enables the user to access a resource or service.
    Type: Grant
    Filed: October 25, 2018
    Date of Patent: August 11, 2020
    Assignee: International Business Machines Corporation
    Inventors: Ayelet Avni, Fady Copty, Ayman Jarrous, Sharon Keidar-Barner, Shiri Lemel
  • Patent number: 10740168
    Abstract: A computer-implemented method, system, and non-transitory computer program product for maintaining a system. A domain of the system is identified. Problem information identifying a problem in the system is captured. Key performance indicators are obtained from a historical model database for the identified domain. An unsupervised model is applied to the key performance indicators to identify historical solutions to historical problems that are similar to the problem in the system. A linear complexity model is used to identify potential solutions for the problem as an historical solution for each historical problem that has the lowest combination of time-cost complexity, resource-cost complexity, and recurrence frequency for the historical problem. A real-time objective function is used to select a solution to the problem from the potential solutions. A maintenance operation to be performed on the system to implement the solution and fix the problem is identified.
    Type: Grant
    Filed: March 29, 2018
    Date of Patent: August 11, 2020
    Assignee: International Business Machines Corporation
    Inventors: Ramshanker Kowta, Ruchi Mahindru, Awadesh Tiwari
  • Patent number: 10740223
    Abstract: A system described herein may use automated techniques, such as machine learning techniques, to identify sequences of actions that satisfy checkpoint criteria. Different sequences of actions may be used for different iterations of the same checkpoints, and may be used to refine a model that evaluates the different sequences of actions (e.g., scores the different sequences of actions). The model may be used to simulate the same or similar actions, in order to validate or discover other sequences of actions for the same checkpoint.
    Type: Grant
    Filed: January 31, 2019
    Date of Patent: August 11, 2020
    Assignee: Verizon Patent and Licensing, Inc.
    Inventor: Qiao Yu
  • Patent number: 10737385
    Abstract: A hand for transferring a transfer target can be selected even when the combination of the transfer target and the hand is not taught. A machine learning device includes: state observation means for acquiring at least a portion of image data obtained by imaging a transfer target as input data; label acquisition means for acquiring information related to grasping means attached to a robot for transferring the transfer target as a label; and learning means for performing supervised learning using a set of the input data acquired by the state observation means and the label acquired by the label acquisition means as teacher data to construct a learning model that outputs information related to the grasping means appropriate for the transferring.
    Type: Grant
    Filed: August 30, 2018
    Date of Patent: August 11, 2020
    Assignee: FANUC CORPORATION
    Inventors: Yasuhiro Shibasaki, Yuta Namiki, Hiromitsu Takahashi
  • Patent number: 10739763
    Abstract: An industrial controller within an industrial automation environment is provided. The industrial controller includes a programming logic controller, configured to control an industrial device, and a predictive module, coupled with the programming logic controller. The predictive module is configured to receive configuration data from the programming logic controller, create a user-defined controller tag, and transfer the user-defined controller tag to the programming logic controller. The predictive module is also configured to receive operational data from the programming logic controller, and calculate operation monitoring data and value estimation data based on the operational data. The predictive module is further configured to write the operation monitoring data and value estimation data to the user-defined controller tag.
    Type: Grant
    Filed: November 13, 2018
    Date of Patent: August 11, 2020
    Assignee: Rockwell Automation Technologies, Inc.
    Inventors: Eugene Mourzine, Alexander B. Smith, Jonathan M. Wise, Olivia M. Leslie, Samuel L. Massari, Jonathan D. Walter
  • Patent number: 10733344
    Abstract: A computer implemented method of selecting a prover among a plurality of provers for a design to be verified. The method comprises collecting, by a data module, raw data relating to the design, and extracting from the raw data a plurality of input features, transforming, by a transformer module, the plurality of input features, wherein transforming the plurality of features comprises applying a linear regression to the plurality of features, classifying using a classification module, the provers from the plurality of provers, in which the classification module is adapted to predict a best prover being the prover which solves a property faster than the remaining provers of the plurality of provers, selecting one or more provers based on the results of the classification.
    Type: Grant
    Filed: October 23, 2017
    Date of Patent: August 4, 2020
    Assignee: Onespin Solutions GmbH
    Inventor: Monica Rafaila
  • Patent number: 10732983
    Abstract: A system including: at least one processor; and at least one memory having stored thereon computer program code that, when executed by the at least one processor, controls the system to: receive a data model identification and a dataset; in response to determining that the data model does not contain a hierarchical structure, perform expectation propagation on the dataset to approximate the data model with a hierarchical structure; divide the dataset into a plurality of channels; for each of the plurality of channels: divide the data into a plurality of microbatches; process each microbatch of the plurality of microbatches through parallel iterators; and process the output of the parallel iterators through single-instruction multiple-data (SIMD) layers; and asynchronously merge results of the SIMD layers.
    Type: Grant
    Filed: May 2, 2019
    Date of Patent: August 4, 2020
    Assignee: CAPITAL ONE SERVICES, LLC
    Inventors: Matthew van Adelsberg, Rohit Joshi, Siqi Wang
  • Patent number: 10735285
    Abstract: Embodiments provide systems and methods for identifying and mitigating outlier network activity. In embodiments, network activity by a plurality of users may be monitored and, based on the monitoring, a plurality of data sets may be compiled. Each of the plurality data sets may include information representative of activity by the plurality of users. A network model representative of at least a portion of the network activity may be constructed based on one or more of the plurality of data sets. The network model may be evaluated against a set of rules to produce outputs that include at least one of: a set of classifications, a set of link metrics, and a set of communities. Decision engine logic may be executed against the outputs to identify outlier network activity. In response to identifying outlier network activity, operations to mitigate the identified outlier network activity may be executed.
    Type: Grant
    Filed: August 9, 2018
    Date of Patent: August 4, 2020
    Assignee: Accenture Global Solutions Limited
    Inventors: Mayank Verma, Loizos Markides, Athina Kanioura, Ray Eitel Porter, Gerasimos Mileounis, Kieran Towey, Mark Ghannam, Vladyslav Yakovenko
  • Patent number: 10733537
    Abstract: A method for ensemble based labeling is provided. The method includes obtaining a plurality of samples of an object. The method further includes estimating, for each of the plurality of samples, a probability that a label applies to the sample, for each of a plurality of labels. The method also includes determining a candidate label among the plurality of labels, based on the estimated probabilities of the plurality of samples for each of the plurality of labels. The method further includes calculating a dispersion of the estimated probabilities of the plurality of samples for the candidate label; and identifying a target label among the plurality of labels, based on the estimated probabilities of the plurality of samples for the candidate label, the dispersion for the candidate label, and a number of the plurality of samples.
    Type: Grant
    Filed: November 9, 2017
    Date of Patent: August 4, 2020
    Assignee: International Business Machines Corporation
    Inventor: Hiroshi Inoue
  • Patent number: 10733080
    Abstract: A static analysis tool configured to determine a significance of static analysis results. The static analysis tool includes computer program code to perform a static analysis of a computer program and generate the static analysis results in response to the performance of the static analysis of the computer program. The program code can further analyze a description of a result item from the static analysis results, and based on the analysis of the description of the result item, assign to the result item information from an ontology scheme. The program code can further include code determine a significance value for the result item in response to the assignment of the information from the ontology scheme and automatically perform an action associated with the result item based on one or more of the information assigned from the ontology scheme or the significance value.
    Type: Grant
    Filed: June 27, 2016
    Date of Patent: August 4, 2020
    Assignee: International Business Machines Corporation
    Inventors: Fionnuala G. Gunter, Christy L. Norman Perez, Michael T. Strosaker, George C. Wilson
  • Patent number: 10732952
    Abstract: This disclosure relates to customizing deployment of an application to a user interface of a client device. An exemplary method generally includes training a model based on historical context information of a plurality of users by identifying correlations between the historical context information and a plurality of widgets and storing the correlations in the model. The method further includes receiving context information from the client device. The method further includes determining a user intent based on the context information using the model. The method further includes selecting one or more widgets to include in a custom user interface definition based, at least in part, on the user intent. The method further includes transmitting, to the user interface of the client device, the custom user interface definition.
    Type: Grant
    Filed: February 6, 2018
    Date of Patent: August 4, 2020
    Assignee: INTUIT, INC.
    Inventors: Jay Yu, Amit Arya, Alexey Povkh, Jeffery Brewer, Elangovan Shanmugam, Gaurav V. Chaubal, Yamit P. Mody
  • Patent number: 10735212
    Abstract: In order to facilitate electronic meeting scheduling and coordination, systems and methods are disclosed including receiving, by a processor, a plurality of electronic meeting requests to schedule a meeting. The processor determines, for each electronic meeting request, meeting room needs. A meeting scheduling machine learning model is utilized to predict parameters of meeting room objects representing the candidate meeting rooms based at least in part on the meeting room needs, schedule information associated with a respective electronic meeting request and location information associated with the respective electronic meeting request. The processor causes an indication of the candidate meeting rooms to display in response to the electronic meeting request on a screen of computing devices associated with the respective attendees based at least in part on the predicted parameters.
    Type: Grant
    Filed: January 21, 2020
    Date of Patent: August 4, 2020
    Assignee: Capital One Services, LLC
    Inventors: James Zarakas, George Bergeron, Adam Vukich
  • Patent number: 10733287
    Abstract: One embodiment provides a method, including: deploying a machine learning model, wherein the deployed machine learning model is used in responding to queries from users; receiving, at the deployed machine learning model, input from a user; identifying a type of machine learning model attack corresponding to the received input; computing, responsive to receiving the input, a resiliency score of the machine learning model, wherein the resiliency score indicates resistance of the machine learning model against the identified type of attack; and performing an action responsive to the computed resiliency score.
    Type: Grant
    Filed: May 14, 2018
    Date of Patent: August 4, 2020
    Assignee: INTERNATIONAL BUSINESS MACHINES CORPORATION
    Inventors: Manish Kesarwani, Suranjana Samanta, Deepak Vijaykeerthy, Sameep Mehta, Karthik Sankaranarayanan
  • Patent number: 10735490
    Abstract: Devices and systems for voice over Internet protocol (VoIP) for identifying network traffic are described herein. One or more embodiments include a VoIP device for identifying network traffic comprising a signal monitor to identify a signaling protocol from the network traffic and an artificial intelligence engine configured to receive signaling protocol sample data to train a signal artificial intelligence (AI) model and process the signaling protocol identified by the signal monitor in the signal AI model to identify the network traffic.
    Type: Grant
    Filed: June 14, 2018
    Date of Patent: August 4, 2020
    Assignee: Edgewater Networks, Inc.
    Inventor: Surendra Prajapat
  • Patent number: 10726297
    Abstract: Systems and methods for identifying semantically and/or visually related information among a set of content items, such content items that include similar concepts or that have similar visual aspects, are disclosed. The disclosed techniques provide tools for identifying related information among various content items, such as text pages and documents, presentation slides and slide decks, etc. The disclosed techniques provide improved methods for searching among content items, organizing content items into categories, and pruning redundant content. Furthermore, the disclosed techniques provide improvements to computation of various metrics, including usage, performance, and impact metrics.
    Type: Grant
    Filed: May 21, 2018
    Date of Patent: July 28, 2020
    Assignee: Highspot, Inc.
    Inventors: Raphael Hoffman, Nate Dire, Erik Christensen, Oliver Sharp, David Wortendyke, Scot Gellock, Robert Wahbe
  • Patent number: 10726025
    Abstract: In an example, a plurality of user profiles in a social networking service are accessed. A heterogeneous graph structure having a plurality of nodes connected by edges is generated, each node corresponding to a different entity in the social networking service, each edge representing a co-occurrence of entities represented by nodes on each side of the edge in at least one of the user profiles. Weights are calculated for each edge of the heterogeneous graph structure, the weights being based on co-occurrence counts reflecting a number of user profiles in the plurality of user profiles in which corresponding nodes co-occurred. The heterogeneous graph structure is embedded into a d-dimensional space. A machine-learned model is then used to calculate a similarity score between a first node and second node by computing distance between the first node and the second node in the d¬-dimensional space.
    Type: Grant
    Filed: February 19, 2018
    Date of Patent: July 28, 2020
    Assignee: Microsoft Technology Licensing, LLC
    Inventors: Qi Guo, Xianren Wu, Bo Hu, Shan Zhou, Lei Ni, Erik Eugene Buchanan
  • Patent number: 10726308
    Abstract: In some examples, image content moderation may include classifying, based on a learning model, an object displayed in an image into a category. Further, image content moderation may include detecting, based on another learning model, the object, refining the detected object based on a label, and determining, based on the another learning model, a category for the refined detected object. Further, image content moderation may include identifying, based on the label, a keyword associated with the object, and determining, based on the identified keyword, a category for the object. Further, image content moderation may include categorizing, based on a set of rules, the object into a category, and moderating image content by categorizing, based on aforementioned analysis the object into a category. Yet further, image content moderation may include tagging, based on fusion-based tagging, the object with a category and a color associated with the object.
    Type: Grant
    Filed: September 26, 2017
    Date of Patent: July 28, 2020
    Assignee: ACCENTURE GLOBAL SOLUTIONS LIMITED
    Inventors: Amioy Kumar, Nagendra K. Kumar, Madhura Shivaram, Suraj Govind Jadhav, Chung-Sheng Li, Saurabh Mahadik
  • Patent number: 10726348
    Abstract: Some embodiments perform probabilistic request routing in addition to or instead of deterministic request routing. The probabilistic request routing is based on probabilistic models that predict the type of content being requested based on commonality in elements between different requests directed to the same type. The probabilistic models accurately route requests that have not been previously encountered and accurately route requests for content whose type is not previously known. The requests are routed across different subsets of servers that are optimized or configured for the predicted type. The probabilistic models can be defined using a decision tree. Machine learning generates and maintains the decision tree. Accuracy predicted by the different branches of the tree is updated through tracking the type of content passed in response to different routed requests. The tree structure is modified based on timestamps associated with the tree elements and based on newly encountered request elements.
    Type: Grant
    Filed: September 20, 2016
    Date of Patent: July 28, 2020
    Assignee: Verizon Digital Media Services Inc.
    Inventors: Hooman Mahyar, Amir Reza Khakpour
  • Patent number: 10725735
    Abstract: Systems and methods for merging annotations of datasets are provided. For example, assignments of labels to data-points may be obtained, confidence levels associated with the assignments of labels may be obtained. Further, the assignments of labels may be merged, for example based on the confidence levels. In some cases, inference models may be generated using the merged assignment of labels. In some examples, an update to the assignments of labels to data-points and/or the confidence levels may be obtained, and the merged assignment of labels may be updated.
    Type: Grant
    Filed: July 30, 2018
    Date of Patent: July 28, 2020
    Assignee: Allegro Artificial Intelligence LTD
    Inventor: Moshe Guttmann
  • Patent number: 10726466
    Abstract: A method, system and a computer program product are provided for making product recommendations to improve a user's personal brand by using the symbolic meanings and utilities of products and a user's brand perceptions along with users input of a desired personal brand imagery to output a set of products that are optimized to help users bridge the gaps between their desired and actual self-brand, thereby enabling a user to navigate products that shape their personal imagery.
    Type: Grant
    Filed: November 3, 2016
    Date of Patent: July 28, 2020
    Assignee: International Business Machines Corporation
    Inventors: Rama K. Akkiraju, Haibin Liu, Jalal U. Mahmud, Vibha S. Sinha, Anbang Xu
  • Patent number: 10719778
    Abstract: A model learning unit of an anomaly detection device learns a relational expression between vibration strengths at frequencies based on a time series of frequency characteristics of a vibration strength detected during a learning period by a vibration sensor placed on a monitoring target. The anomaly detection unit learns a relational expression between vibration strengths at frequencies based on a time series of frequency characteristics of a vibration strength detected during a new period by the vibration sensor. Then, the anomaly detection unit determines whether or not there is an anomaly in the monitoring target based on a relational expression related to a new frequency, which is different from the relational expression learned during the learning period.
    Type: Grant
    Filed: May 11, 2015
    Date of Patent: July 21, 2020
    Assignee: NEC CORPORATION
    Inventor: Katsuhiro Ochiai
  • Patent number: 10719826
    Abstract: A method, apparatus, and system for paying are provided. The method for paying of an electronic device includes: receiving item order information from a POS terminal; determining a payment method by considering at least one of a discount benefit and a reward point saving benefit based on the received item order information; and requesting a payment approval for the item order information in the determined payment method.
    Type: Grant
    Filed: March 13, 2015
    Date of Patent: July 21, 2020
    Assignee: Samsung Electronics Co., Ltd.
    Inventors: Ho Kim, Jin-Wan Choi, Soo-Bin Park
  • Patent number: 10713539
    Abstract: A non-transitory recording medium stores a program causing a computer to execute a process including determining a reference value on a decision surface classifying a group of image data represented by a plurality of types of feature quantities into two classes in a feature quantity space, based on a feature quantity of given image data that is closest to the decision surface and is classified into one of the classes of the group of image data in the feature quantity space, plotting and displaying pieces of image data of the group of image data along at least one axis indicating a degree of separation of each of the pieces of image data relative to the reference value, modifying, upon receiving a single piece of image data identified from the plotted and displayed pieces of image data, the decision surface based on the identified single piece of image data.
    Type: Grant
    Filed: May 23, 2018
    Date of Patent: July 14, 2020
    Assignee: FUJITSU LIMITED
    Inventors: Hiroaki Takebe, Masaki Ishihara, Masahiko Sugimura, Susumu Endo, Takayuki Baba, Yusuke Uehara
  • Patent number: 10710239
    Abstract: Systems, computer program products, and methods are described herein for intelligent control code update for robotic process automation. The present invention is configured to retrieve execution logs associated with robotic process automation (RPA) sessions, wherein the execution logs comprises exceptions. Next, the present invention is configured to initiate machine learning algorithms configured to process the one or more execution logs and classify the exceptions into predetermined classes. Next, the present invention is configured to deploy automated exception handling subroutines to address the exceptions based on at least classifying the exceptions into the predetermined classes.
    Type: Grant
    Filed: November 8, 2018
    Date of Patent: July 14, 2020
    Assignee: Bank of America Corporation
    Inventors: Jigesh Rajendra Safary, Krishna Rangarao Mamadapur, Gyanendra Mohan Sinha
  • Patent number: 10713769
    Abstract: Methods and systems for performing active learning for defect classifiers are provided. One system includes one or more computer subsystems configured for performing active learning for training a defect classifier. The active learning includes applying an acquisition function to data points for the specimen. The acquisition function selects one or more of the data points based on uncertainty estimations associated with the data points. The active learning also includes acquiring labels for the selected one or more data points and generating a set of labeled data that includes the selected one or more data points and the acquired labels. The computer subsystem(s) are also configured for training the defect classifier using the set of labeled data. The defect classifier is configured for classifying defects detected on the specimen using the images generated by the imaging subsystem.
    Type: Grant
    Filed: May 28, 2019
    Date of Patent: July 14, 2020
    Assignee: KLA-Tencor Corp.
    Inventors: Jing Zhang, Yujie Dong, Brian Duffy, Richard Wallingford, Michael Daino, Kris Bhaskar
  • Patent number: 10713384
    Abstract: A relational database is transformed so as to obfuscate secure and/or private aspects of data contained in the database, while preserving salient elements of the data to facilitate data analysis. A restructured database is generatively modeled, and the model is sampled to create synthetic data that maintains sufficiently similar (or the same) mathematical properties and relations as the original data stored in the database. In one example, various statistics at the intersection of related database tables are determined by modeling data using an iterative multivariate approach. Synthetic data may be sampled from any part of the modeled database, wherein the synthesized data is “realistic” in that it statistically mimics the original data in the database. The generation of such synthetic data allows publication of bulk data freely and on-demand (e.g., for data analysis purposes), without the risk of security/privacy breaches.
    Type: Grant
    Filed: December 8, 2017
    Date of Patent: July 14, 2020
    Assignees: Massachusetts Institute of Technology, ACCENTURE GLOBAL SOLUTIONS LIMITED
    Inventors: Kalyan Kumar Veeramachaneni, Neha Patki, Kishore Prabhakar Durg, Jeffrey Steven Wilkinson, Sunder Ranganathan Nochilur
  • Patent number: 10705506
    Abstract: A machine learning device performs reinforcement learning on a controller that performs multiple processes for controlling a machine tool in parallel at multiple operation units.
    Type: Grant
    Filed: May 16, 2018
    Date of Patent: July 7, 2020
    Assignee: FANUC CORPORATION
    Inventor: Akira Kanemaru
  • Patent number: 10706365
    Abstract: Techniques facilitating local optimization of quantum circuits are provided. In one example, a computer-implemented method comprises applying, by a device operatively coupled to a processor, respective weights to matrix elements of a first matrix corresponding to a quantum circuit according to respective numbers of quantum gates between respective pairs of qubits in the quantum circuit; transforming, by the device, the first matrix into a second matrix based on the respective weights of the matrix elements; and permuting, by the device, respective qubits in the quantum circuit according to the second matrix, resulting in a permuted quantum circuit.
    Type: Grant
    Filed: September 27, 2018
    Date of Patent: July 7, 2020
    Assignee: INTERNATIONAL BUSINESS MACHINES CORPORATION
    Inventor: Paul Nation
  • Patent number: 10706328
    Abstract: A method is described for generating a prediction of a disease classification error for a magnified, digital microscope slide image of a tissue sample. The image is composed of a multitude of patches or tiles of pixel image data. An out-of-focus degree per patch is computed using a machine learning out-of-focus classifier. Data representing expected disease classifier error statistics of a machine learning disease classifier for a plurality of out-of-focus degrees is retrieved. A mapping of the expected disease classifier error statistics to each of the patches of the digital microscope slide image based on the computed out-of-focus degree per patch is computed, thereby generating a disease classifier error prediction for each of the patches. The disease classifier error predictions thus generated are aggregated over all of the patches.
    Type: Grant
    Filed: May 7, 2018
    Date of Patent: July 7, 2020
    Assignee: Google LLC
    Inventors: Martin Stumpe, Timo Kohlberger
  • Patent number: 10706287
    Abstract: The invention relates to behavioral experiments with rodents, and especially to monitoring, data collection and control of such experiments. This invention provides a method for automatically observing the behavior of rodents in a way that provides accurate determination of location and direction of the head of the rodent, even in darkness under infrared light, by determining the location and the direction of the rodent's head on the basis of the triangle formed by the ears and the snout of the rodent.
    Type: Grant
    Filed: April 21, 2017
    Date of Patent: July 7, 2020
    Inventor: Petri Ala-Laurila
  • Patent number: 10705979
    Abstract: An apparatus, method, program product, and system are disclosed for evicting pages from memory using a neural network. One embodiment of a method for evicting pages from memory using a neural network includes determining state information related to evicting pages from memory. The state information may be determined by a dedicated hardware snooping device that snoops a system bus for the state information. The method includes determining an identifier for a page in memory to be evicted using a neural network. The neural network performs machine learning operations on the state information to identify the page in memory to be evicted. The method includes locating the identified page in memory using the identifier determined by the neural network and evicting the identified page from memory.
    Type: Grant
    Filed: October 27, 2017
    Date of Patent: July 7, 2020
    Assignee: INTERNATIONAL BUSINESS MACHINES CORPORATION
    Inventors: Amanda A. Liem, Matthew R. Ochs, Lennard G. Streat, Brendan M. Wong
  • Patent number: 10699803
    Abstract: Presenting ancestral origin information, comprising: receiving a request to display ancestry data of an individual; obtaining ancestry composition information of the individual, the ancestry composition information including information pertaining to a proportion of the individual's genotype data that is deemed to correspond to a specific ancestry; and presenting the ancestry composition information to be displayed.
    Type: Grant
    Filed: June 13, 2016
    Date of Patent: June 30, 2020
    Assignee: 23andMe, Inc.
    Inventors: Chuong Do, Eric Durand, John Michael Macpherson
  • Patent number: 10700992
    Abstract: A method of managing resources in a cloud environment is disclosed. The method includes receiving a plurality of parameters associated with an event. The method further includes comparing a value of each of the plurality of parameters with a predefined threshold range. The method includes converting the value of each of the plurality of parameters into a vector, when the value of each of the plurality of parameters is within the predefined threshold range. The method further includes training a neural network based on the vector of the value of each of the plurality of parameters, wherein the neural network is trained to manage the event. The method includes storing an output of the trained neural network in a database in response to the training. The output corresponds to management of the event and the database further comprises a mapping of the event to the trained neural network.
    Type: Grant
    Filed: March 26, 2019
    Date of Patent: June 30, 2020
    Assignee: Wipro Limited
    Inventors: Rishav Das, Maulik Yagnik
  • Patent number: 10701085
    Abstract: Communication partners known to be malignant or benign are input to a known communication partner input unit, a subject communication partner whose malignancy is to be calculated is input to a subject communication partner input unit, a characteristic extractor extracts changes over time in whether the known communication partners and the subject communication partner are listed at a past given time point on a malignancy communication partner list and a benign communication partner list, and a malignancy calculator calculates malignancy of the subject communication partner on the basis of the characteristic information about the known communication partners and the subject communication partner.
    Type: Grant
    Filed: February 12, 2016
    Date of Patent: June 30, 2020
    Assignee: NIPPON TELEGRAPH AND TELEPHONE CORPORATION
    Inventors: Daiki Chiba, Takeshi Yagi
  • Patent number: 10699233
    Abstract: Systems, methods, and other embodiments associated with dynamic predictive modelling. According to one embodiment, a method includes creating a joined pair including a snapshot time and the forecast time. The joined pair is stored in the storage device. A subset of data associated with the joined pair is selected from data stored in the storage device. The subset of data is selected based, at least in part, on predetermined increments of time between the snapshot time and the forecast time. The snapshot time, the forecast time, and the subset of data is provided to a processing device from the storage device.
    Type: Grant
    Filed: July 22, 2015
    Date of Patent: June 30, 2020
    Assignee: WELLS FARGO BANK, N.A.
    Inventors: Jie Chen, Weicheng Liu
  • Patent number: 10691952
    Abstract: A method of tracking a position of a target object in a video sequence includes identifying the target object in a reference frame. A generic mapping is applied to the target object being tracked. The generic mapping is generated by learning possible appearance variations of a generic object. The method also includes tracking the position of the target object in subsequent frames of the video sequence by determining whether an output of the generic mapping of the target object matches an output of the generic mapping of a candidate object.
    Type: Grant
    Filed: September 20, 2019
    Date of Patent: June 23, 2020
    Assignee: QUALCOMM Incorporated
    Inventors: Ran Tao, Efstratios Gavves, Arnold Wilhelmus Maria Smeulders
  • Patent number: 10693907
    Abstract: Disclosed are a system, a method, and computer readable storage medium having instructions for filtering network traffic to protect a server from a distributed denial-of-service (DDoS) attack. The described technique includes intercepting data from a network node to the computing device responsive to detecting a computing device is subject to a DDoS attack. The technique further includes determining one or more data transmission parameters based on the intercepted data, assigning a danger rating to the network node, and changing the danger rating of the network node based on application of a filter and on the data transmission parameters. The described technique limits a transmittal of data from the network node to the computing device if the resultant danger rating of the network node exceeds a threshold value.
    Type: Grant
    Filed: June 6, 2017
    Date of Patent: June 23, 2020
    Assignee: AO Kaspersky Lab
    Inventors: Nikolay V. Gudov, Alexander A. Khalimonenko, Denis E. Koreshkov
  • Patent number: 10691850
    Abstract: A power analysis system for an integrated circuit device design can use machine learning to determine an estimated power consumption of the design. In various examples, the system can generate workloads for a power projection tool, which can include less than all the data of a full suite of power projection tests. The results from the power projection tool can be used to train a machine learning data model. From the results, the data model can learn the functions of the design by grouping together cells that are triggered together by the same signals. The data model can also learn estimated power consumption for each of the functions. The output of the data model can then be used to configure a design testing tool, which can run tests on the design. The output of the tests can then be used to compute an estimated overall power consumption for the design.
    Type: Grant
    Filed: December 13, 2018
    Date of Patent: June 23, 2020
    Assignee: Amazon Technologies, Inc.
    Inventors: Lev Makovsky, Adi Habusha, Ron Diamant
  • Patent number: 10693740
    Abstract: A device may receive one or more data models that have been trained using a first set of values that are in a format capable of being processed by the one or more data models. The first set of values may be associated with a set of historical network performance indicators relating to a set of network devices. The device may receive network data that includes network ticket information and performance statistics for the one or more network devices. The device may determine a set of network performance indicators relating to the one or more network devices. The device may convert the set of network performance indicators into a second set of values that are in the format capable of being processed by the one or more data models. The device may use the second set of values to generate one or more recommendations associated with improving network performance.
    Type: Grant
    Filed: December 7, 2017
    Date of Patent: June 23, 2020
    Assignee: Accenture Global Solutions Limited
    Inventors: Davide Coccia, Davide Guglielmo Bellini
  • Patent number: 10692006
    Abstract: A chatbot can use a knowledge base including question/answer pairs to respond to questions. When a question is asked that does not correspond to a question/answer pair in the knowledge base, the chatbot can send the question to one or more humans to obtain an answer. However, only some people will have the experience, context, knowledge, etc., to answer the question. A model can be trained to select “experts” that are likely to be able to provide a good answer to a question by using both A) a vector comprising characteristics of questions and of the person posing the questions and B) a vector comprising characteristics of a possible expert. The model can trained to produce a value predicting how good an identified expert's answer is likely to be. The model can be trained based on measures of past answers provided for types of questions/questioners.
    Type: Grant
    Filed: June 30, 2016
    Date of Patent: June 23, 2020
    Assignee: FACEBOOK, INC.
    Inventor: Ying Zhang
  • Patent number: 10691890
    Abstract: A word segmentation method and system for a language text, where in the method, a word segmentation is performed on the first language text in a first word segmentation manner to obtain a first word boundary set, the first word boundary set is divided into a trusted second word boundary set and an untrusted third word boundary set according to a confidence level threshold, a second language text is selected from the first language text according to the third word boundary set, and a word segmentation is performed on the second language text in a second word segmentation manner to obtain a fourth word boundary set. Word segmentation precision of the first language text can be flexibly adjusted by adjusting the confidence level threshold.
    Type: Grant
    Filed: September 18, 2018
    Date of Patent: June 23, 2020
    Assignee: HUAWEI TECHNOLOGIES CO., LTD.
    Inventors: Xiao Chen, Hang Li
  • Patent number: 10692015
    Abstract: A method and a machine learning relationship determination system (MLRDS) for determining primary key-foreign key (PK-FK) relationships among data in tables of a target database through machine learning (ML) are provided. The MLRDS selects columns of the tables in the target database and identifies inclusion dependency (ID) pairs from the selected columns. The MLRDS receives training data and validation data from a source database, computes PK-FK features for the inclusion dependency pairs, the training data, and the validation data, and generates trained ML models and validated ML models using the PK-FK features. The MLRDS determines an optimum algorithm decision threshold for a selected machine learning classification algorithm (MLCA), using which the MLRDS determines a resultant on whether the inclusion dependency pair is a PK-FK pair or a non-PK-FK pair. The MLRDS performs majority voting on the resultant for multiple MLCAs to confirm the PK-FK relationships between the inclusion dependency pairs.
    Type: Grant
    Filed: July 15, 2016
    Date of Patent: June 23, 2020
    Assignee: Io-Tahoe LLC
    Inventors: Yongming Xu, Ram Dayal Goyal
  • Patent number: 10684950
    Abstract: Embodiments of the present invention provide a system for triggering cross channel data caching. Historical event data and live event data of a user may be monitored to determine an expected event that comprises one or more expected channels. An expected period of time for the expected event may also be determined. Relevant user data may then be identified from one or more systems of record and cached or otherwise compressed. One or more adapters configured to format cached data into the one or more expected channels are then identified. The cached data is then transmitted to databases associated with each of the one or more expected channels, where the cached data is converted by the adapters. The cached data is then generally maintained in these databases for the expected period of time.
    Type: Grant
    Filed: March 15, 2018
    Date of Patent: June 16, 2020
    Assignee: Bank of America Corporation
    Inventors: Sandeep Kumar Chauhan, Sarat Kumar Magatapalli
  • Patent number: 10685260
    Abstract: Systems and methods are disclosed that enable distributed execution of prediction models by disparate, remote systems. Prediction model code is transmitted to the disparate, distributed systems for execution by the disparate, remote systems. Default model input data may be independently modified by a given system, and the modified input data may be used when the given system executes the model. Model predictions and associated model parameters are received from the disparate, distributed systems. The accuracy of the received model predictions from the disparate, distributed systems are analyzed. Based on the analyzed accuracy of the received model predictions, a determination is made as to which model predictions satisfy at least a first criterion. Computer-based resources are allocated using the determination as to which model predictions satisfy at least the first criterion.
    Type: Grant
    Filed: June 6, 2019
    Date of Patent: June 16, 2020
    Assignee: Finiti Research Limited
    Inventors: Jesse David Adelaar, Werner Janjic, Christoph Giess
  • Patent number: 10684290
    Abstract: Described herein are methods and compositions for the diagnosis, prognosis, selection of treatment and treatment of cancer, and particularly, of lung cancer such as non-small cell lung cancer. Embodiments of the present invention involve the detection of LKB1 levels and sensitivity to endoplasmic reticulum (ER) stress. Treatment can be made through the administration of ER stress activators.
    Type: Grant
    Filed: February 18, 2015
    Date of Patent: June 16, 2020
    Assignee: Dignity Health
    Inventor: Landon J. Inge
  • Patent number: 10686774
    Abstract: An authentication system comprises a client device system associated with a session user, a data provider server system that stores user data, including user identification data, associated with the session user, a data provider interface system for displaying a data provider user interface on the client device system, an authentication server system that stores authentication data associated with the session user, and an authentication interface system for displaying an authentication user interface on the client device system.
    Type: Grant
    Filed: January 16, 2018
    Date of Patent: June 16, 2020
    Assignee: ASIGNIO INC.
    Inventors: Kyle Rutherford, Eric Dustrude, Erik Hodge, Benjamin MacKay, Calvin Rutherford, Kevin Boyd