Patents by Inventor Frank Torsten Bernd Seide

Frank Torsten Bernd Seide has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).

  • Patent number: 11200269
    Abstract: Examples of the present disclosure describe systems and methods relating to generating relevance scores for one or more words of a passage which is an answer to a natural language query. For instance, a passage extracted from a highly relevant electronic file along with the query may encoded and augmented to generate a multi-dimensional, augmented semantic vectors using recurring neural networks. The augmented semantic vectors along with a multi-dimensional vector that represent words of the passage may be decoded to generate relevance scores for one or more words of the passage, based on levels of relevance to the query.
    Type: Grant
    Filed: June 15, 2017
    Date of Patent: December 14, 2021
    Assignee: Microsoft Technology Licensing, LLC
    Inventors: Qifa Ke, Frank Torsten Bernd Seide, Qi Liu, Rajanala Sai Krishna Sravanthi
  • Patent number: 11049006
    Abstract: Techniques and constructs can reduce the time required to determine solutions to optimization problems such as training of neural networks. Modifications to a computational model can be determined by a plurality of nodes operating in parallel. Quantized modification values can be transmitted between the nodes to reduce the volume of data to be transferred. The quantized values can be as small as one bit each. Quantization-error values can be stored and used in quantizing subsequent modifications. The nodes can operate in parallel and overlap computation and data transfer to further reduce the time required to determine solutions. The quantized values can be partitioned and each node can aggregate values for a corresponding partition.
    Type: Grant
    Filed: September 12, 2014
    Date of Patent: June 29, 2021
    Assignee: Microsoft Technology Licensing, LLC
    Inventors: John Langford, Gang Li, Frank Torsten Bernd Seide, James Droppo, Dong Yu
  • Patent number: 10956535
    Abstract: Disclosed in some examples are methods, systems, machine-readable media, and devices which operate a neural network defined by user code. A method includes identifying, operations from user code that are integral in operating the neural network, combining a subset of the identified operations into a single processing sequence to be transmitted to an array of hardware processors, performing operations that are not integral in operation of the neural network in a separate thread of execution from the operations that are integral in operating the neural network; and mapping results to the combined operations that were included in the single processing sequence.
    Type: Grant
    Filed: June 15, 2017
    Date of Patent: March 23, 2021
    Assignee: MICROSOFT TECHNOLOGY LICENSING, LLC
    Inventors: Frank Torsten Bernd Seide, Ryota Tomioka, Wilhelm Richert, Bruno S Bozza
  • Patent number: 10325200
    Abstract: Discriminative pretraining technique embodiments are presented that pretrain the hidden layers of a Deep Neural Network (DNN). In general, a one-hidden-layer neural network is trained first using labels discriminatively with error back-propagation (BP). Then, after discarding an output layer in the previous one-hidden-layer neural network, another randomly initialized hidden layer is added on top of the previously trained hidden layer along with a new output layer that represents the targets for classification or recognition. The resulting multiple-hidden-layer DNN is then discriminatively trained using the same strategy, and so on until the desired number of hidden layers is reached. This produces a pretrained DNN. The discriminative pretraining technique embodiments have the advantage of bringing the DNN layer weights close to a good local optimum, while still leaving them in a range with a high gradient so that they can be fine-tuned effectively.
    Type: Grant
    Filed: October 1, 2015
    Date of Patent: June 18, 2019
    Assignee: MICROSOFT TECHNOLOGY LICENSING, LLC
    Inventors: Dong Yu, Li Deng, Frank Torsten Bernd Seide, Gang Li
  • Publication number: 20190065954
    Abstract: In a data center, neural network evaluations can be included for services involving image or speech recognition by using a field programmable gate array (FPGA) or other parallel processor. The memory bandwidth limitations of providing weighted data sets from an external memory to the FPGA (or other parallel processor) can be managed by queuing up input data from the plurality of cores executing the services at the FPGA (or other parallel processor) in batches of at least two feature vectors. The at least two feature vectors can be at least two observation vectors from a same data stream or from different data streams. The FPGA (or other parallel processor) can then act on the batch of data for each loading of the weighted datasets.
    Type: Application
    Filed: October 30, 2018
    Publication date: February 28, 2019
    Inventors: Ray A. BITTNER, JR., Frank Torsten Bernd SEIDE
  • Publication number: 20180365321
    Abstract: Examples of the present disclosure describe systems and methods relating to generating relevance scores for one or more words of a passage which is an answer to a natural language query. For instance, a passage extracted from a highly relevant electronic file along with the query may encoded and augmented to generate a multi-dimensional, augmented semantic vectors using recurring neural networks. The augmented semantic vectors along with a multi-dimensional vector that represent words of the passage may be decoded to generate relevance scores for one or more words of the passage, based on levels of relevance to the query.
    Type: Application
    Filed: June 15, 2017
    Publication date: December 20, 2018
    Applicant: Microsoft Technology Licensing, LLC
    Inventors: Qifa KE, Frank Torsten Bernd SEIDE, Qi LIU, Rajanala Sai Krishna SRAVANTHI
  • Patent number: 10140572
    Abstract: In a data center, neural network evaluations can be included for services involving image or speech recognition by using a field programmable gate array (FPGA) or other parallel processor. The memory bandwidth limitations of providing weighted data sets from an external memory to the FPGA (or other parallel processor) can be managed by queuing up input data from the plurality of cores executing the services at the FPGA (or other parallel processor) in batches of at least two feature vectors. The at least two feature vectors can be at least two observation vectors from a same data stream or from different data streams. The FPGA (or other parallel processor) can then act on the batch of data for each loading of the weighted datasets.
    Type: Grant
    Filed: June 25, 2015
    Date of Patent: November 27, 2018
    Assignee: Microsoft Technology Licensing, LLC
    Inventors: Ray A. Bittner, Jr., Frank Torsten Bernd Seide
  • Publication number: 20180336461
    Abstract: Disclosed in some examples are methods, systems, machine-readable media, and devices which operate a neural network defined by user code. A method includes identifying, operations from user code that are integral in operating the neural network, combining a subset of the identified operations into a single processing sequence to be transmitted to an array of hardware processors, performing operations that are not integral in operation of the neural network in a separate thread of execution from the operations that are integral in operating the neural network; and mapping results to the combined operations that were included in the single processing sequence.
    Type: Application
    Filed: June 15, 2017
    Publication date: November 22, 2018
    Inventors: FRANK TORSTEN BERND SEIDE, RYOTA TOMIOKA, WILHELM RICHERT, BRUNO S. BOZZA
  • Patent number: 10095694
    Abstract: Content-based analysis is performed on multimedia content prior to encoding the multimedia content in the rendering chain of processing. A content-based index stream is generated based on the content-based analysis and the content-based index stream is embedded in the multimedia file during rendering. The content-based index stream can be used to generate a content-based searchable index when necessary.
    Type: Grant
    Filed: May 12, 2016
    Date of Patent: October 9, 2018
    Assignee: Microsoft Technology Licensing, LLC
    Inventors: Albert J. K. Thambiratnam, Frank Torsten Bernd Seide, Roger Peng Yu
  • Publication number: 20170308789
    Abstract: Techniques and constructs can reduce the time required to determine solutions to optimization problems such as training of neural networks. Modifications to a computational model can be determined by a plurality of nodes operating in parallel. Quantized modification values can be transmitted between the nodes to reduce the volume of data to be transferred. The quantized values can be as small as one bit each. Quantization-error values can be stored and used in quantizing subsequent modifications. The nodes can operate in parallel and overlap computation and data transfer to further reduce the time required to determine solutions. The quantized values can be partitioned and each node can aggregate values for a corresponding partition.
    Type: Application
    Filed: September 12, 2014
    Publication date: October 26, 2017
    Inventors: John LANGFORD, Gang LI, Frank Torsten Bernd SEIDE, James DROPPO, Dong YU
  • Publication number: 20160379111
    Abstract: In a data center, neural network evaluations can be included for services involving image or speech recognition by using a field programmable gate array (FPGA) or other parallel processor. The memory bandwidth limitations of providing weighted data sets from an external memory to the FPGA (or other parallel processor) can be managed by queuing up input data from the plurality of cores executing the services at the FPGA (or other parallel processor) in batches of at least two feature vectors. The at least two feature vectors can be at least two observation vectors from a same data stream or from different data streams. The FPGA (or other parallel processor) can then act on the batch of data for each loading of the weighted datasets.
    Type: Application
    Filed: June 25, 2015
    Publication date: December 29, 2016
    Inventors: Ray A. BITTNER, JR., Frank Torsten Bernd SEIDE
  • Patent number: 9495449
    Abstract: Described is a technology by which a playback list comprising similar songs is automatically built based on automatically detected/generated song attributes, such as by extracting numeric features of each song. The attributes may be downloaded from a remote connection, and/or may be locally generated on the playback device. To build a playlist, a seed song's attributes may be compared against attributes of other songs to determine which other songs are similar to the seed song and thus included in the playlist. Another way to build a playlist is based on similarity of songs to a set of user provided-attributes, such as corresponding to moods or usage modes such as “resting” “reading” “jogging” or “driving” moods/modes. The playlist may be dynamically adjusted based on user interaction with the device, such as when a user skips a song, queues a song, or dequeues a song.
    Type: Grant
    Filed: February 3, 2014
    Date of Patent: November 15, 2016
    Assignee: Microsoft Technology Licensing, LLC
    Inventors: Lie Lu, Frank Torsten Bernd Seide, Gabriel White
  • Patent number: 9483557
    Abstract: In various embodiments, a transcript that represents a media file is created. Keyword candidates that may represent topics and/or content associated with the media content are then be extracted from the transcript. Furthermore, a keyword set may be generated for the media content utilizing a mutual information criteria. In other embodiments, one or more queries may be generated based at least in part on the transcript, and a plurality of web documents may be retrieved based at least in part on the one or more queries. Additional keyword candidates may be extracted from each web document and then ranked. A subset of the keyword candidates may then be selected to form a keyword set associated with the media content.
    Type: Grant
    Filed: March 4, 2011
    Date of Patent: November 1, 2016
    Assignee: Microsoft Technology Licensing LLC
    Inventors: Albert Joseph Kishan Thambiratnam, Sha Meng, Gang Li, Frank Torsten Bernd Seide
  • Patent number: 9477925
    Abstract: The use of a pipelined algorithm that performs parallelized computations to train deep neural networks (DNNs) for performing data analysis may reduce training time. The DNNs may be one of context-independent DNNs or context-dependent DNNs. The training may include partitioning training data into sample batches of a specific batch size. The partitioning may be performed based on rates of data transfers between processors that execute the pipelined algorithm, considerations of accuracy and convergence, and the execution speed of each processor. Other techniques for training may include grouping layers of the DNNs for processing on a single processor, distributing a layer of the DNNs to multiple processors for processing, or modifying an execution order of steps in the pipelined algorithm.
    Type: Grant
    Filed: November 20, 2012
    Date of Patent: October 25, 2016
    Assignee: Microsoft Technology Licensing, LLC
    Inventors: Frank Torsten Bernd Seide, Gang Li, Dong Yu, Adam C. Eversole, Xie Chen
  • Publication number: 20160267177
    Abstract: Described is a technology by which a playback list comprising similar songs is automatically built based on automatically detected/generated song attributes, such as by extracting numeric features of each song. The attributes may be downloaded from a remote connection, and/or may be locally generated on the playback device. To build a playlist, a seed song's attributes may be compared against attributes of other songs to determine which other songs are similar to the seed song and thus included in the playlist. Another way to build a playlist is based on similarity of songs to a set of user provided-attributes, such as corresponding to moods or usage modes such as “resting” “reading” “jogging” or “driving” moods/modes. The playlist may be dynamically adjusted based on user interaction with the device, such as when a user skips a song, queues a song, or dequeues a song.
    Type: Application
    Filed: May 25, 2016
    Publication date: September 15, 2016
    Inventors: Lie Lu, Frank Torsten Bernd Seide, Gabriel White
  • Publication number: 20160259782
    Abstract: Content-based analysis is performed on multimedia content prior to encoding the multimedia content in the rendering chain of processing. A content-based index stream is generated based on the content-based analysis and the content-based index stream is embedded in the multimedia file during rendering. The content-based index stream can be used to generate a content-based searchable index when necessary.
    Type: Application
    Filed: May 12, 2016
    Publication date: September 8, 2016
    Inventors: Albert J.K. Thambiratnam, Frank Torsten Bernd Seide, Roger Peng Yu
  • Publication number: 20160026914
    Abstract: Discriminative pretraining technique embodiments are presented that pretrain the hidden layers of a Deep Neural Network (DNN). In general, a one-hidden-layer neural network is trained first using labels discriminatively with error back-propagation (BP). Then, after discarding an output layer in the previous one-hidden-layer neural network, another randomly initialized hidden layer is added on top of the previously trained hidden layer along with a new output layer that represents the targets for classification or recognition. The resulting multiple-hidden-layer DNN is then discriminatively trained using the same strategy, and so on until the desired number of hidden layers is reached. This produces a pretrained DNN. The discriminative pretraining technique embodiments have the advantage of bringing the DNN layer weights close to a good local optimum, while still leaving them in a range with a high gradient so that they can be fine-tuned effectively.
    Type: Application
    Filed: October 1, 2015
    Publication date: January 28, 2016
    Inventors: Dong Yu, Li Deng, Frank Torsten Bernd Seide, Gang Li
  • Patent number: 9235799
    Abstract: Discriminative pretraining technique embodiments are presented that pretrain the hidden layers of a Deep Neural Network (DNN). In general, a one-hidden-layer neural network is trained first using labels discriminatively with error back-propagation (BP). Then, after discarding an output layer in the previous one-hidden-layer neural network, another randomly initialized hidden layer is added on top of the previously trained hidden layer along with a new output layer that represents the targets for classification or recognition. The resulting multiple-hidden-layer DNN is then discriminatively trained using the same strategy, and so on until the desired number of hidden layers is reached. This produces a pretrained DNN. The discriminative pretraining technique embodiments have the advantage of bringing the DNN layer weights close to a good local optimum, while still leaving them in a range with a high gradient so that they can be fine-tuned effectively.
    Type: Grant
    Filed: November 26, 2011
    Date of Patent: January 12, 2016
    Assignee: MICROSOFT TECHNOLOGY LICENSING, LLC
    Inventors: Dong Yu, Li Deng, Frank Torsten Bernd Seide, Gang Li
  • Patent number: 8825481
    Abstract: Techniques are described for training a speech recognition model for accented speech. A subword parse table is employed that models mispronunciations at multiple subword levels, such as the syllable, position-specific cluster, and/or phone levels. Mispronunciation probability data is then generated at each level based on inputted training data, such as phone-level annotated transcripts of accented speech. Data from different levels of the subword parse table may then be combined to determine the accented speech model. Mispronunciation probability data at each subword level is based at least in part on context at that level. In some embodiments, phone-level annotated transcripts are generated using a semi-supervised method.
    Type: Grant
    Filed: January 20, 2012
    Date of Patent: September 2, 2014
    Assignee: Microsoft Corporation
    Inventors: Albert Joseph Kishan Thambiratnam, Timo Pascal Mertens, Frank Torsten Bernd Seide
  • Publication number: 20140149468
    Abstract: Described is a technology by which a playback list comprising similar songs is automatically built based on automatically detected/generated song attributes, such as by extracting numeric features of each song. The attributes may be downloaded from a remote connection, and/or may be locally generated on the playback device. To build a playlist, a seed song's attributes may be compared against attributes of other songs to determine which other songs are similar to the seed song and thus included in the playlist. Another way to build a playlist is based on similarity of songs to a set of user provided-attributes, such as corresponding to moods or usage modes such as “resting” “reading” “jogging” or “driving” moods/modes. The playlist may be dynamically adjusted based on user interaction with the device, such as when a user skips a song, queues a song, or dequeues a song.
    Type: Application
    Filed: February 3, 2014
    Publication date: May 29, 2014
    Applicant: MICROSOFT CORPORATION
    Inventors: Lie Lu, Frank Torsten Bernd Seide, Gabriel White