Patents by Inventor Daniel Stefan Tarlow
Daniel Stefan Tarlow 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).
-
Publication number: 20240036832Abstract: A computer program predictor is described which has a processor configured to access a program attribute predictor; and a memory storing a search component configured to search a space of possible programs, to find a program which, given an input data instance and an output data instance, will compute the output data instance from the input data instance, the search being guided by attributes predicted by the attribute predictor given the input data instance and the output data instance.Type: ApplicationFiled: October 9, 2023Publication date: February 1, 2024Inventors: Alexander Lloyd GAUNT, Sebastian NOWOZIN, Marc Manuel Johannes BROCKSCHMIDT, Daniel Stefan TARLOW, Matej BALOG
-
Publication number: 20230393817Abstract: Systems and methods of the present disclosure are directed to a method for machine-learned code segment prediction for optimizing software development. The method includes obtaining an incomplete segment of code. The method includes processing the incomplete segment of code with a machine-learned code prediction model to obtain a sampled set of segment completion predictions that include code that completes the incomplete segment of code. The method includes determining an aggregated segment completion prediction from the sampled set of segment completion predictions. The method includes replacing a portion of the aggregated segment completion prediction with an input field, wherein the portion of the aggregated segment completion prediction is associated with a degree of certainty less than a threshold degree of certainty.Type: ApplicationFiled: June 3, 2022Publication date: December 7, 2023Inventors: Daniel Dun-ning Woo Johnson, Daniel Stefan Tarlow, Maxim Tabachnyk, Marc Hatcher Rasi, Jacob Austin, Hassan Abolhassani, Jacob Hanson Hegna
-
Patent number: 11816457Abstract: A computer program predictor is described which has a processor configured to access a program attribute predictor; and a memory storing a search component configured to search a space of possible programs, to find a program which, given an input data instance and an output data instance, will compute the output data instance from the input data instance, the search being guided by attributes predicted by the attribute predictor given the input data instance and the output data instance.Type: GrantFiled: August 28, 2020Date of Patent: November 14, 2023Assignee: Microsoft Technology Licensing, LLCInventors: Alexander Lloyd Gaunt, Sebastian Nowozin, Marc Manuel Johannes Brockschmidt, Daniel Stefan Tarlow, Matej Balog
-
Publication number: 20220222531Abstract: A neural network training apparatus is described which has a network of worker nodes each having a memory storing a subgraph of a neural network to be trained. The apparatus has a control node connected to the network of worker nodes. The control node is configured to send training data instances into the network to trigger parallelized message passing operations which implement a training algorithm which trains the neural network. At least some of the message passing operations asynchronously update parameters of individual subgraphs of the neural network at the individual worker nodes.Type: ApplicationFiled: March 28, 2022Publication date: July 14, 2022Inventors: Ryota TOMIOKA, Matthew Alastair JOHNSON, Daniel Stefan TARLOW, Samuel Alexander WEBSTER, Dimitrios VYTINIOTIS, Alexander Lloyd GAUNT, Maik RIECHERT
-
Patent number: 11288575Abstract: A neural network training apparatus is described which has a network of worker nodes each having a memory storing a subgraph of a neural network to be trained. The apparatus has a control node connected to the network of worker nodes. The control node is configured to send training data instances into the network to trigger parallelized message passing operations which implement a training algorithm which trains the neural network. At least some of the message passing operations asynchronously update parameters of individual subgraphs of the neural network at the individual worker nodes.Type: GrantFiled: May 18, 2017Date of Patent: March 29, 2022Assignee: Microsoft Technology Licensing, LLCInventors: Ryota Tomioka, Matthew Alastair Johnson, Daniel Stefan Tarlow, Samuel Alexander Webster, Dimitrios Vytiniotis, Alexander Lloyd Gaunt, Maik Riechert
-
Publication number: 20200394024Abstract: A computer program predictor is described which has a processor configured to access a program attribute predictor; and a memory storing a search component configured to search a space of possible programs, to find a program which, given an input data instance and an output data instance, will compute the output data instance from the input data instance, the search being guided by attributes predicted by the attribute predictor given the input data instance and the output data instance.Type: ApplicationFiled: August 28, 2020Publication date: December 17, 2020Inventors: Alexander Lloyd GAUNT, Sebastian NOWOZIN, Marc Manuel Johannes BROCKSCHMIDT, Daniel Stefan TARLOW, Matej BALOG
-
Patent number: 10782939Abstract: A computer program predictor is described which has a processor configured to access a program attribute predictor; and a memory storing a search component configured to search a space of possible programs, to find a program which, given an input data instance and an output data instance, will compute the output data instance from the input data instance, the search being guided by attributes predicted by the attribute predictor given the input data instance and the output data instance.Type: GrantFiled: August 7, 2017Date of Patent: September 22, 2020Assignee: Microsoft Technology Licensing, LLCInventors: Alexander Lloyd Gaunt, Sebastian Nowozin, Marc Manuel Johannes Brockschmidt, Daniel Stefan Tarlow, Matej Balog
-
Publication number: 20190042210Abstract: A computer program predictor is described which has a processor configured to access a program attribute predictor; and a memory storing a search component configured to search a space of possible programs, to find a program which, given an input data instance and an output data instance, will compute the output data instance from the input data instance, the search being guided by attributes predicted by the attribute predictor given the input data instance and the output data instance.Type: ApplicationFiled: August 7, 2017Publication date: February 7, 2019Inventors: Alexander Lloyd GAUNT, Sebastian NOWOZIN, Marc Manuel Johannes BROCKSCHMIDT, Daniel Stefan TARLOW, Matej BALOG
-
Publication number: 20180336458Abstract: A neural network training apparatus is described which has a network of worker nodes each having a memory storing a subgraph of a neural network to be trained. The apparatus has a control node connected to the network of worker nodes. The control node is configured to send training data instances into the network to trigger parallelized message passing operations which implement a training algorithm which trains the neural network. At least some of the message passing operations asynchronously update parameters of individual subgraphs of the neural network at the individual worker nodes.Type: ApplicationFiled: May 18, 2017Publication date: November 22, 2018Inventors: Ryota TOMIOKA, Matthew Alastair JOHNSON, Daniel Stefan TARLOW, Samuel Alexander WEBSTER, Dimitrios VYTINIOTIS, Alexander Lloyd GAUNT, Maik RIECHERT
-
Patent number: 10127497Abstract: An inference engine is described for efficient machine learning. For example, an inference engine executes a plurality of ordered steps to carry out inference on the basis of observed data. For each step, a plurality of inputs to the step are received. A predictor predicts an output of the step and computes uncertainty of the prediction. Either the predicted output or a known output is selected on the basis of the uncertainty. If the known output is selected, the known output is computed, (for example, using a resource intensive, accurate process). The predictor is retrained using the known output and the plurality of inputs of the step as training data. For example, computing the prediction is fast and efficient as compared with computing the known output.Type: GrantFiled: October 14, 2014Date of Patent: November 13, 2018Assignee: Microsoft Technology Licensing, LLCInventors: Seyed Mohammadali Eslami, Daniel Stefan Tarlow, Pushmeet Kohli, John Winn
-
Patent number: 10037624Abstract: Examples describe an apparatus for calibrating a three dimensional (3D) mesh model of an articulated object. The articulated object is an instance of a specified object class. The apparatus comprises an input configured to receive captured sensor data depicting the object. The apparatus has a calibration engine configured to compute values of shape parameters of the 3D mesh model which indicate which member of the object class is depicted in the captured sensor data, in order to calibrate the 3D mesh model. The calibration engine is configured to compute the values of the shape parameters with an optimization process to find at least one potential local or global minimum of an energy function, the energy function expressing a degree of similarity between data rendered from the model and the received sensor data.Type: GrantFiled: December 29, 2015Date of Patent: July 31, 2018Assignee: Microsoft Technology Licensing, LLCInventors: Thomas Joseph Cashman, David Joseph New Tan, Jamie Daniel Joseph Shotton, Andrew William Fitzgibbon, Sameh Khamis, Jonathan James Taylor, Toby Sharp, Daniel Stefan Tarlow
-
Patent number: 9928040Abstract: Automated generation, or completion, or checking of source code is described whereby a probabilistic model having been trained using a corpus of natural source code examples is used. In various examples the probabilistic model comprises probability distributions describing belief about structure of natural source code and takes into account source code analysis from a compiler or other source code analyzer. In various examples, source code analysis may comprise syntactic structure, type information and other data about source code. In various examples, the trained probabilistic model is used to predict sequences of source code elements. For example, to generate source code, to auto-complete source code, to error check source code, to error correct source code or for other purposes.Type: GrantFiled: February 26, 2014Date of Patent: March 27, 2018Assignee: Microsoft Technology Licensing, LLCInventors: Daniel Stefan Tarlow, Christopher Joseph Maddison
-
Publication number: 20170186226Abstract: Examples describe an apparatus for calibrating a three dimensional (3D) mesh model of an articulated object. The articulated object is an instance of a specified object class. The apparatus comprises an input configured to receive captured sensor data depicting the object. The apparatus has a calibration engine configured to compute values of shape parameters of the 3D mesh model which indicate which member of the object class is depicted in the captured sensor data, in order to calibrate the 3D mesh model. The calibration engine is configured to compute the values of the shape parameters with an optimization process to find at least one potential local or global minimum of an energy function, the energy function expressing a degree of similarity between data rendered from the model and the received sensor data.Type: ApplicationFiled: December 29, 2015Publication date: June 29, 2017Inventors: Thomas Joseph CASHMAN, David Joseph New TAN, Jamie Daniel Joseph SHOTTON, Andrew William FITZGIBBON, Sameh KHAMIS, Jonathan James TAYLOR, Toby SHARP, Daniel Stefan TARLOW
-
Publication number: 20160104070Abstract: An inference engine is described for efficient machine learning. For example, an inference engine executes a plurality of ordered steps to carry out inference on the basis of observed data. For each step, a plurality of inputs to the step are received. A predictor predicts an output of the step and computes uncertainty of the prediction. Either the predicted output or a known output is selected on the basis of the uncertainty. If the known output is selected, the known output is computed, (for example, using a resource intensive, accurate process). The predictor is retrained using the known output and the plurality of inputs of the step as training data. For example, computing the prediction is fast and efficient as compared with computing the known output.Type: ApplicationFiled: October 14, 2014Publication date: April 14, 2016Inventors: Seyed Mohammadali Eslami, Daniel Stefan Tarlow, Pushmeet Kohli, John Winn
-
Publication number: 20150135166Abstract: Automated generation, or completion, or checking of source code is described whereby a probabilistic model having been trained using a corpus of natural source code examples is used. In various examples the probabilistic model comprises probability distributions describing belief about structure of natural source code and takes into account source code analysis from a compiler or other source code analyzer. In various examples, source code analysis may comprise syntactic structure, type information and other data about source code. In various examples, the trained probabilistic model is used to predict sequences of source code elements. For example, to generate source code, to auto-complete source code, to error check source code, to error correct source code or for other purposes.Type: ApplicationFiled: February 26, 2014Publication date: May 14, 2015Applicant: Microsoft CorporationInventors: Daniel Stefan Tarlow, Christopher Joseph Maddison