Patents by Inventor Georgios Evangelopoulos
Georgios Evangelopoulos 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: 20240077848Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for an exosuit activity transition control structure. In some implementations, sensor data for a powered exosuit is received. The sensor data is classified depending on whether the sensor data is indicative of a transition between different types of activities of a wearer of the powered exosuit. The classification is provided to a control system for the powered exosuit. The powered exosuit is controlled based on the classification.Type: ApplicationFiled: November 9, 2023Publication date: March 7, 2024Applicant: Skip Innovations, Inc.Inventors: Kathryn Jane Zealand, Elliott J. Rouse, Georgios Evangelopoulos
-
Patent number: 11853034Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for an exosuit activity transition control structure. In some implementations, sensor data for a powered exosuit is received. The sensor data is classified depending on whether the sensor data is indicative of a transition between different types of activities of a wearer of the powered exosuit. The classification is provided to a control system for the powered exosuit. The powered exosuit is controlled based on the classification.Type: GrantFiled: December 3, 2020Date of Patent: December 26, 2023Assignee: Skip Innovations, Inc.Inventors: Kathryn Jane Zealand, Elliott J. Rouse, Georgios Evangelopoulos
-
Patent number: 11748065Abstract: Techniques are described herein for using artificial intelligence to “learn,” statistically, a target programming style that is imposed in and/or evidenced by a code base. Once the target programming style is learned, it can be used for various purposes. In various implementations, one or more generative adversarial networks (“GANs”), each including a generator machine learning model and a discriminator machine learning model, may be trained to facilitate learning and application of target programming style(s). In some implementations, the discriminator(s) and/or generator(s) may operate on graphical input, and may take the form of graph neural networks (“GNNs”), graph attention neural networks (“GANNs”), graph convolutional networks (“GCNs”), etc., although this is not required.Type: GrantFiled: December 28, 2021Date of Patent: September 5, 2023Assignee: GOOGLE LLCInventors: Georgios Evangelopoulos, Olivia Hatalsky, Bin Ni, Qianyu Zhang
-
Publication number: 20230229891Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for implementing a reservoir computing neural network. In one aspect there is provided a reservoir computing neural network comprising: (i) a brain emulation sub-network, and (ii) a prediction sub-network. The brain emulation sub-network is configured to process the network input in accordance with values of a plurality of brain emulation sub-network parameters to generate an alternative representation of the network input. The prediction sub-network is configured to process the alternative representation of the network input in accordance with values of a plurality of prediction sub-network parameters to generate the network output. The values of the brain emulation sub-network parameters are determined before the reservoir computing neural network is trained and are not adjusting during training of the reservoir computing neural network.Type: ApplicationFiled: February 23, 2023Publication date: July 20, 2023Inventors: Sarah Ann Laszlo, Philip Edwin Watson, Georgios Evangelopoulos
-
Patent number: 11620487Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for selecting a neural network architecture for performing a machine learning task. In one aspect, a method comprises: obtaining data defining a synaptic connectivity graph representing synaptic connectivity between neurons in a brain of a biological organism; generating data defining a plurality of candidate graphs based on the synaptic connectivity graph; determining, for each candidate graph, a performance measure on a machine learning task of a neural network having a neural network architecture that is specified by the candidate graph; and selecting a final neural network architecture for performing the machine learning task based on the performance measures.Type: GrantFiled: January 29, 2020Date of Patent: April 4, 2023Assignee: X Development LLCInventors: Sarah Ann Laszlo, Philip Edwin Watson, Georgios Evangelopoulos
-
Patent number: 11593617Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for implementing a reservoir computing neural network. In one aspect there is provided a reservoir computing neural network comprising: (i) a brain emulation sub-network, and (ii) a prediction sub-network. The brain emulation sub-network is configured to process the network input in accordance with values of a plurality of brain emulation sub-network parameters to generate an alternative representation of the network input. The prediction sub-network is configured to process the alternative representation of the network input in accordance with values of a plurality of prediction sub-network parameters to generate the network output. The values of the brain emulation sub-network parameters are determined before the reservoir computing neural network is trained and are not adjusting during training of the reservoir computing neural network.Type: GrantFiled: January 30, 2020Date of Patent: February 28, 2023Assignee: X Development LLCInventors: Sarah Ann Laszlo, Philip Edwin Watson, Georgios Evangelopoulos
-
Patent number: 11568201Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for determining an artificial neural network architecture corresponding to a sub-graph of a synaptic connectivity graph. In one aspect, there is provided a method comprising: obtaining data defining a graph representing synaptic connectivity between neurons in a brain of a biological organism; determining, for each node in the graph, a respective set of one or more node features characterizing a structure of the graph relative to the node; identifying a sub-graph of the graph, comprising selecting a proper subset of the nodes in the graph for inclusion in the sub-graph based on the node features of the nodes in the graph; and determining an artificial neural network architecture corresponding to the sub-graph of the graph.Type: GrantFiled: January 30, 2020Date of Patent: January 31, 2023Assignee: X Development LLCInventors: Sarah Ann Laszlo, Georgios Evangelopoulos, Philip Edwin Watson
-
Publication number: 20220121427Abstract: Techniques are described herein for using artificial intelligence to “learn,” statistically, a target programming style that is imposed in and/or evidenced by a code base. Once the target programming style is learned, it can be used for various purposes. In various implementations, one or more generative adversarial networks (“GANs”), each including a generator machine learning model and a discriminator machine learning model, may be trained to facilitate learning and application of target programming style(s). In some implementations, the discriminator(s) and/or generator(s) may operate on graphical input, and may take the form of graph neural networks (“GNNs”), graph attention neural networks (“GANNs”), graph convolutional networks (“GCNs”), etc., although this is not required.Type: ApplicationFiled: December 28, 2021Publication date: April 21, 2022Inventors: Georgios Evangelopoulos, Olivia Hatalsky, Bin Ni, Qianyu Zhang
-
Publication number: 20220096249Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for an exosuit activity transition control structure. In some implementations, sensor data representing an estimate of sensor data that will be produced by sensors of an exosuit at a particular time in the future is generated, where the exosuit is configured to assist mobility of a wearer. Actual sensor data that is generated using the sensors of the exosuit is obtained. A measure of uncertainty is determined based on the forecasted sensor data and the actual sensor data. Based on the measure of uncertainty, a control action for the exosuit to adjust an assistance provided to the wearer is determined.Type: ApplicationFiled: July 30, 2021Publication date: March 31, 2022Inventors: Elliott J. Rouse, Georgios Evangelopoulos, Kathryn Jane Zealand
-
Patent number: 11243746Abstract: Techniques are described herein for using artificial intelligence to “learn,” statistically, a target programming style that is imposed in and/or evidenced by a code base. Once the target programming style is learned, it can be used for various purposes. In various implementations, one or more generative adversarial networks (“GANs”), each including a generator machine learning model and a discriminator machine learning model, may be trained to facilitate learning and application of target programming style(s). In some implementations, the discriminator(s) and/or generator(s) may operate on graphical input, and may take the form of graph neural networks (“GNNs”), graph attention neural networks (“GANNs”), graph convolutional networks (“GCNs”), etc., although this is not required.Type: GrantFiled: July 1, 2019Date of Patent: February 8, 2022Assignee: X DEVELOPMENT LLCInventors: Georgios Evangelopoulos, Olivia Hatalsky, Bin Ni, Qianyu Zhang
-
Publication number: 20220004167Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for generating, using, or both, exosuit historical data. In some implementations, (i) sensor data generated by sensors of an exosuit worn by a user and (ii) control data indicating actions performed by or control signals generated by the exosuit based on the sensor data while worn by the user are received. The sensor data and the control data are added to a database that includes historical data describing use of the exosuit over time by the user. A control scheme of the exosuit is customized for the user by updating the one or more machine learning models or settings that govern the application of the one or more machine learning models. Forces provided by one or more actuators of the exosuit are controlled using the updated one or more machine learning models or the updated settings.Type: ApplicationFiled: July 2, 2021Publication date: January 6, 2022Inventors: Kathryn Jane Zealand, Joseph Hollis Sargent, Georgios Evangelopoulos, Elliott J. Rouse
-
Publication number: 20210349445Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for an exosuit activity transition control structure. In some implementations, sensor data for a powered exosuit is received. The sensor data is classified depending on whether the sensor data is indicative of a transition between different types of activities of a wearer of the powered exosuit. The classification is provided to a control system for the powered exosuit. The powered exosuit is controlled based on the classification.Type: ApplicationFiled: December 3, 2020Publication date: November 11, 2021Inventors: Kathryn Jane Zealand, Elliott J. Rouse, Georgios Evangelopoulos
-
Patent number: 11169786Abstract: Implementations are described herein for generating embeddings of source code using both the language and graph domains, and leveraging combinations of these semantically-rich and structurally-informative embeddings for various purposes. In various implementations, tokens of a source code snippet may be applied as input across a sequence-processing machine learning model to generate a plurality of token embeddings. A graph may also be generated based on the source code snippet. A joint representation may be generated based on the graph and the incorporated token embeddings. The joint representation generated from the source code snippet may be compared to one or more other joint representations generated from one or more other source code snippets to make a determination about the source code snippet.Type: GrantFiled: February 4, 2020Date of Patent: November 9, 2021Assignee: X DEVELOPMENT LLCInventors: Rohan Badlani, Owen Lewis, Georgios Evangelopoulos, Olivia Hatalsky, Bin Ni
-
Publication number: 20210240453Abstract: Implementations are described herein for generating embeddings of source code using both the language and graph domains, and leveraging combinations of these semantically-rich and structurally-informative embeddings for various purposes. In various implementations, tokens of a source code snippet may be applied as input across a sequence-processing machine learning model to generate a plurality of token embeddings. A graph may also be generated based on the source code snippet. A joint representation may be generated based on the graph and the incorporated token embeddings. The joint representation generated from the source code snippet may be compared to one or more other joint representations generated from one or more other source code snippets to make a determination about the source code snippet.Type: ApplicationFiled: February 4, 2020Publication date: August 5, 2021Inventors: Rohan Badlani, Owen Lewis, Georgios Evangelopoulos, Olivia Hatalsky, Bin Ni
-
Publication number: 20210201111Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for determining an artificial neural network architecture corresponding to a sub-graph of a synaptic connectivity graph. In one aspect, there is provided a method comprising: obtaining data defining a graph representing synaptic connectivity between neurons in a brain of a biological organism; determining, for each node in the graph, a respective set of one or more node features characterizing a structure of the graph relative to the node; identifying a sub-graph of the graph, comprising selecting a proper subset of the nodes in the graph for inclusion in the sub-graph based on the node features of the nodes in the graph; and determining an artificial neural network architecture corresponding to the sub-graph of the graph.Type: ApplicationFiled: January 30, 2020Publication date: July 1, 2021Inventors: Sarah Ann Laszlo, Georgios Evangelopoulos, Philip Edwin Watson
-
Publication number: 20210201115Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for implementing a reservoir computing neural network. In one aspect there is provided a reservoir computing neural network comprising: (i) a brain emulation sub-network, and (ii) a prediction sub-network. The brain emulation sub-network is configured to process the network input in accordance with values of a plurality of brain emulation sub-network parameters to generate an alternative representation of the network input. The prediction sub-network is configured to process the alternative representation of the network input in accordance with values of a plurality of prediction sub-network parameters to generate the network output. The values of the brain emulation sub-network parameters are determined before the reservoir computing neural network is trained and are not adjusting during training of the reservoir computing neural network.Type: ApplicationFiled: January 30, 2020Publication date: July 1, 2021Inventors: Sarah Ann Laszlo, Philip Edwin Watson, Georgios Evangelopoulos
-
Publication number: 20210201107Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for selecting a neural network architecture for performing a machine learning task. In one aspect, a method comprises: obtaining data defining a synaptic connectivity graph representing synaptic connectivity between neurons in a brain of a biological organism; generating data defining a plurality of candidate graphs based on the synaptic connectivity graph; determining, for each candidate graph, a performance measure on a machine learning task of a neural network having a neural network architecture that is specified by the candidate graph; and selecting a final neural network architecture for performing the machine learning task based on the performance measures.Type: ApplicationFiled: January 29, 2020Publication date: July 1, 2021Inventors: Sarah Ann Laszlo, Philip Edwin Watson, Georgios Evangelopoulos
-
Patent number: 11048482Abstract: Implementations are described herein for automatically identifying, recommending, and/or automatically effecting changes to a source code base based on updates previously made to other similar code bases. Intuitively, multiple prior “migrations,” or mass updates, of complex software system code bases may be analyzed to identify changes that were made. More particularly, a particular portion or “snippet” of source code—which may include a whole source code file, a source code function, a portion of source code, or any other semantically-meaningful code unit—may undergo a sequence of edits over time. Techniques described herein leverage this sequence of edits to predict a next edit of the source code snippet. These techniques have a wide variety of applications, including but not limited to automatically updating of source code, source code completion, recommending changes to source code, etc.Type: GrantFiled: July 26, 2019Date of Patent: June 29, 2021Assignee: X DEVELOPMENT LLCInventors: Georgios Evangelopoulos, Benoit Schillings, Bin Ni
-
Publication number: 20210026605Abstract: Implementations are described herein for automatically identifying, recommending, and/or automatically effecting changes to a source code base based on updates previously made to other similar code bases. Intuitively, multiple prior “migrations,” or mass updates, of complex software system code bases may be analyzed to identify changes that were made. More particularly, a particular portion or “snippet” of source code—which may include a whole source code file, a source code function, a portion of source code, or any other semantically-meaningful code unit—may undergo a sequence of edits over time. Techniques described herein leverage this sequence of edits to predict a next edit of the source code snippet. These techniques have a wide variety of applications, including but not limited to automatically updating of source code, source code completion, recommending changes to source code, etc.Type: ApplicationFiled: July 26, 2019Publication date: January 28, 2021Inventors: Georgios Evangelopoulos, Benoit Schillings, Bin Ni
-
Publication number: 20210004210Abstract: Techniques are described herein for using artificial intelligence to “learn,” statistically, a target programming style that is imposed in and/or evidenced by a code base. Once the target programming style is learned, it can be used for various purposes. In various implementations, one or more generative adversarial networks (“GANs”), each including a generator machine learning model and a discriminator machine learning model, may be trained to facilitate learning and application of target programming style(s). In some implementations, the discriminator(s) and/or generator(s) may operate on graphical input, and may take the form of graph neural networks (“GNNs”), graph attention neural networks (“GANNs”), graph convolutional networks (“GCNs”), etc., although this is not required.Type: ApplicationFiled: July 1, 2019Publication date: January 7, 2021Inventors: Georgios Evangelopoulos, Olivia Hatalsky, Bin Ni, Qianyu Zhang