Patents by Inventor Yiwen Guo
Yiwen Guo 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: 20240086693Abstract: Methods and systems for budgeted and simplified training of deep neural networks (DNNs) are disclosed. In one example, a trainer is to train a DNN using a plurality of training sub-images derived from a down-sampled training image. A tester is to test the trained DNN using a plurality of testing sub-images derived from a down-sampled testing image. In another example, in a recurrent deep Q-network (RDQN) having a local attention mechanism located between a convolutional neural network (CNN) and a long-short time memory (LSTM), a plurality of feature maps are generated by the CNN from an input image. Hard-attention is applied by the local attention mechanism to the generated plurality of feature maps by selecting a subset of the generated feature maps. Soft attention is applied by the local attention mechanism to the selected subset of generated feature maps by providing weights to the selected subset of generated feature maps in obtaining weighted feature maps.Type: ApplicationFiled: September 22, 2023Publication date: March 14, 2024Inventors: Yiwen GUO, Yuqing Hou, Anbang YAO, Dongqi Cai, Lin Xu, Ping Hu, Shandong Wang, Wenhua Cheng, Yurong Chen, Libin Wang
-
Patent number: 11907843Abstract: Systems, apparatuses and methods may provide for conducting an importance measurement of a plurality of parameters in a trained neural network and setting a subset of the plurality of parameters to zero based on the importance measurement. Additionally, the pruned neural network may be re-trained. In one example, conducting the importance measurement includes comparing two or more parameter values that contain covariance matrix information.Type: GrantFiled: June 30, 2016Date of Patent: February 20, 2024Assignee: Intel CorporationInventors: Anbang Yao, Yiwen Guo, Yurong Chen
-
Patent number: 11887001Abstract: An apparatus and method are described for reducing the parameter density of a deep neural network (DNN). A layer-wise pruning module to prune a specified set of parameters from each layer of a reference dense neural network model to generate a second neural network model having a relatively higher sparsity rate than the reference neural network model; a retraining module to retrain the second neural network model in accordance with a set of training data to generate a retrained second neural network model; and the retraining module to output the retrained second neural network model as a final neural network model if a target sparsity rate has been reached or to provide the retrained second neural network model to the layer-wise pruning model for additional pruning if the target sparsity rate has not been reached.Type: GrantFiled: September 26, 2016Date of Patent: January 30, 2024Assignee: INTEL CORPORATIONInventors: Anbang Yao, Yiwen Guo, Lin Xu, Yan Lin, Yurong Chen
-
Publication number: 20230359873Abstract: Methods and apparatus for discrimitive semantic transfer and physics-inspired optimization in deep learning are disclosed. A computation training method for a convolutional neural network (CNN) includes receiving a sequence of training images in the CNN of a first stage to describe objects of a cluttered scene as a semantic segmentation mask. The semantic segmentation mask is received in a semantic segmentation network of a second stage to produce semantic features. Using weights from the first stage as feature extractors and weights from the second stage as classifiers, edges of the cluttered scene are identified using the semantic features.Type: ApplicationFiled: May 3, 2023Publication date: November 9, 2023Inventors: Anbang YAO, Hao ZHAO, Ming LU, Yiwen GUO, Yurong CHEN
-
Patent number: 11803739Abstract: Methods and systems for budgeted and simplified training of deep neural networks (DNNs) are disclosed. In one example, a trainer is to train a DNN using a plurality of training sub-images derived from a down-sampled training image. A tester is to test the trained DNN using a plurality of testing sub-images derived from a down-sampled testing image. In another example, in a recurrent deep Q-network (RDQN) having a local attention mechanism located between a convolutional neural network (CNN) and a long-short time memory (LSTM), a plurality of feature maps are generated by the CNN from an input image. Hard-attention is applied by the local attention mechanism to the generated plurality of feature maps by selecting a subset of the generated feature maps. Soft attention is applied by the local attention mechanism to the selected subset of generated feature maps by providing weights to the selected subset of generated feature maps in obtaining weighted feature maps.Type: GrantFiled: January 25, 2022Date of Patent: October 31, 2023Assignee: Intel CorporationInventors: Yiwen Guo, Yuqing Hou, Anbang Yao, Dongqi Cai, Lin Xu, Ping Hu, Shandong Wang, Wenhua Cheng, Yurong Chen, Libin Wang
-
Patent number: 11798412Abstract: The disclosure provides a method and device for generating driving suggestion, and computer-readable storage medium. The method comprises: acquiring N driving records, wherein the N driving records are derived from at least two vehicles, each driving record comprises a mapping relationship between a driving period and an acceleration value, and the N is an integer greater than 1; determining a plurality of acceleration metric values based on the acceleration values in the N driving records, wherein each vehicle corresponds to at least one acceleration metric value, and the acceleration metric value is positively correlated with the acceleration value; determining a metric threshold according to the plurality of acceleration metric values; and generating a driving suggestion based on the metric threshold and the driving record corresponding to any one of the at least two vehicles.Type: GrantFiled: June 2, 2021Date of Patent: October 24, 2023Assignee: GUANGZHOU AUTOMOBILE GROUP CO., LTD.Inventors: Yiwen Guo, Wenqi Yang, Rongbin Lin, Yonggang Xu
-
Patent number: 11798191Abstract: A sensor calibrator comprising one or more processors configured to receive sensor data representing a calibration pattern detected by a sensor during a period of relative motion between the sensor and the calibration pattern in which the sensor or the calibration pattern move along a linear path of travel; determine a calibration adjustment from the plurality of images; and send a calibration instruction for calibration of the sensor according to the determined calibration adjustment. Alternatively, a sensor calibration detection device, comprising one or more processors, configured to receive first sensor data detected during movement of a first sensor along a route of travel; determine a difference between the first sensor data and stored second sensor data; and if the difference is outside of a predetermined range, switch from a first operational mode to a second operational mode.Type: GrantFiled: March 27, 2020Date of Patent: October 24, 2023Assignee: Intel CorporationInventors: Ignacio Alvarez, Cornelius Buerkle, Maik Sven Fox, Florian Geissler, Ralf Graefe, Yiwen Guo, Yuqing Hou, Fabian Oboril, Daniel Pohl, Alexander Carl Unnervik, Xiangbin Wu
-
Patent number: 11790223Abstract: Methods and systems are disclosed for boosting deep neural networks for deep learning. In one example, in a deep neural network including a first shallow network and a second shallow network, a first training sample is processed by the first shallow network using equal weights. A loss for the first shallow network is determined based on the processed training sample using equal weights. Weights for the second shallow network are adjusted based on the determined loss for the first shallow network. A second training sample is processed by the second shallow network using the adjusted weights. In another example, in a deep neural network including a first weak network and a second weak network, a first subset of training samples is processed by the first weak network using initialized weights. A classification error for the first weak network on the first subset of training samples is determined.Type: GrantFiled: April 7, 2017Date of Patent: October 17, 2023Assignee: Intel CorporationInventors: Libin Wang, Yiwen Guo, Anbang Yao, Dongqi Cai, Lin Xu, Ping Hu, Shandong Wang, Wenhua Cheng, Yurong Chen
-
Patent number: 11704569Abstract: Methods and apparatus are disclosed for enhancing a binary weight neural network using a dependency tree. A method of enhancing a convolutional neural network (CNN) having binary weights includes constructing a tree for obtained binary tensors, the tree having a plurality of nodes beginning with a root node in each layer of the CNN. A convolution is calculated of an input feature map with an input binary tensor at the root node of the tree. A next node is searched from the root node of the tree and a convolution is calculated at the next node using a previous convolution result calculated at the root node of the tree. The searching of a next node from root node is repeated for all nodes from the root node of the tree, and a convolution is calculated at each next node using a previous convolution result.Type: GrantFiled: May 23, 2018Date of Patent: July 18, 2023Assignee: Intel CorporationInventors: Yiwen Guo, Anbang Yao, Hao Zhao, Ming Lu, Yurong Chen
-
Patent number: 11669718Abstract: Methods and apparatus for discrimitive semantic transfer and physics-inspired optimization in deep learning are disclosed. A computation training method for a convolutional neural network (CNN) includes receiving a sequence of training images in the CNN of a first stage to describe objects of a cluttered scene as a semantic segmentation mask. The semantic segmentation mask is received in a semantic segmentation network of a second stage to produce semantic features. Using weights from the first stage as feature extractors and weights from the second stage as classifiers, edges of the cluttered scene are identified using the semantic features.Type: GrantFiled: May 22, 2018Date of Patent: June 6, 2023Assignee: Intel CorporationInventors: Anbang Yao, Hao Zhao, Ming Lu, Yiwen Guo, Yurong Chen
-
Patent number: 11640526Abstract: Methods and apparatus are disclosed for enhancing a neural network using binary tensor and scale factor pairs. For one example, a method of optimizing a trained convolutional neural network (CNN) includes initializing an approximation residue as a trained weight tensor for the trained CNN. A plurality of binary tensors and scale factor pairs are determined. The approximation residue is updated using the binary tensors and scale factor pairs.Type: GrantFiled: May 22, 2018Date of Patent: May 2, 2023Assignee: Intel CorporationInventors: Yiwen Guo, Anbang Yao, Hao Zhao, Ming Lu, Yurong Chen
-
Patent number: 11635943Abstract: Described herein are hardware acceleration of random number generation for machine learning and deep learning applications. An apparatus (700) includes a uniform random number generator (URNG) circuit (710) to generate uniform random numbers and an adder circuit (750) that is coupled to the URNG circuit (710). The adder circuit hardware (750) accelerates generation of Gaussian random numbers for machine learning.Type: GrantFiled: April 7, 2017Date of Patent: April 25, 2023Assignee: Intel CorporationInventors: Yiwen Guo, Anbang Yao, Dongqi Cai, Libin Wang, Lin Xu, Ping Hu, Shandong Wang, Wenhua Cheng
-
Patent number: 11551335Abstract: Methods and systems are disclosed using camera devices for deep channel and Convolutional Neural Network (CNN) images and formats. In one example, image values are captured by a color sensor array in an image capturing device or camera. The image values provide color channel data. The captured image values by the color sensor array are input to a CNN having at least one CNN layer. The CNN provides CNN channel data for each layer. The color channel data and CNN channel data is to form a deep channel image that stored in a memory. In another example, image values are captured by sensor array. The captured image values by the sensor array are input a CNN having a first CNN layer. An output is generated at the first CNN layer using the captured image values by the color sensor array. The output of the first CNN layer is stored as a feature map of the captured image.Type: GrantFiled: April 7, 2017Date of Patent: January 10, 2023Assignee: Intel CorporationInventors: Lin Xu, Liu Yang, Anbang Yao, Dongqi Cai, Libin Wang, Ping Hu, Shandong Wang, Wenhua Cheng, Yiwen Guo, Yurong Chen
-
Patent number: 11537851Abstract: Methods and systems are disclosed using improved training and learning for deep neural networks. In one example, a deep neural network includes a plurality of layers, and each layer has a plurality of nodes. The nodes of each L layer in the plurality of layers are randomly connected to nodes of an L+1 layer. The nodes of each L+1 layer are connected to nodes in a subsequent L layer in a one-to-one manner. Parameters related to the nodes of each L layer are fixed. Parameters related to the nodes of each L+1 layers are updated. In another example, inputs for the input layer and labels for the output layer of a deep neural network are determined related to a first sample. A similarity between different pairs of inputs and labels is estimated using a Gaussian regression process.Type: GrantFiled: April 7, 2017Date of Patent: December 27, 2022Assignee: Intel CorporationInventors: Yiwen Guo, Anbang Yao, Dongqi Cai, Libin Wang, Lin Xu, Ping Hu, Shandong Wang, Wenhua Cheng, Yurong Chen
-
Publication number: 20220392339Abstract: The disclosure provides a method and device for generating driving suggestion, and computer-readable storage medium. The method comprises: acquiring N driving records, wherein the N driving records are derived from at least two vehicles, each driving record comprises a mapping relationship between a driving period and an acceleration value, and the N is an integer greater than 1; determining a plurality of acceleration metric values based on the acceleration values in the N driving records, wherein each vehicle corresponds to at least one acceleration metric value, and the acceleration metric value is positively correlated with the acceleration value; determining a metric threshold according to the plurality of acceleration metric values; and generating a driving suggestion based on the metric threshold and the driving record corresponding to any one of the at least two vehicles.Type: ApplicationFiled: June 2, 2021Publication date: December 8, 2022Inventors: YIWEN GUO, WENQI YANG, RONGBIN LIN, YONGGANG XU
-
Publication number: 20220374745Abstract: A method for scoring driving behavior using vehicle inputs and outputs is implemented in an electronic device. The method includes obtaining historical input data and output data of a vehicle; establishing an output regression model according to the historical output data; determining a boundary of the output regression model; establishing an input regression model according to the historical input data; determining a boundary of the input regression model by calculating boundary limits of the input regression model; obtaining real-time input data and output data of the vehicle; calculating a first ratio of data points outside the boundary of the input regression model to total data points in the real-time input data, and a second ratio of data points outside the boundary of the output regression model to total data points in the real-time output data; scoring driving behavior of a driver according to the first ratio and the second ratio.Type: ApplicationFiled: May 18, 2021Publication date: November 24, 2022Inventors: KILSOO KIM, Yiwen GUO, Wenqi YANG, Yonggang XU
-
Publication number: 20220230268Abstract: Described herein are advanced artificial intelligence agents for modeling physical interactions. In one embodiment, an apparatus to provide an active artificial intelligence (AI) agent includes at least one database to store physical interaction data and compute cluster coupled to the at least one database. The compute cluster automatically obtains physical interaction data from a data collection module without manual interaction, stores the physical interaction data in the at least one database, and automatically trains diverse sets of machine learning program units to simulate physical interactions with each individual program unit having a different model based on the applied physical interaction data.Type: ApplicationFiled: November 2, 2021Publication date: July 21, 2022Inventors: Anbang YAO, Dongqi CAI, Libin WANG, Lin XU, Ping HU, Shandong WANG, Wenhua CHENG, Yiwen GUO, Liu YANG, Yuqing HOU, Zhou SU
-
Publication number: 20220222492Abstract: Methods and systems for budgeted and simplified training of deep neural networks (DNNs) are disclosed. In one example, a trainer is to train a DNN using a plurality of training sub-images derived from a down-sampled training image. A tester is to test the trained DNN using a plurality of testing sub-images derived from a down-sampled testing image. In another example, in a recurrent deep Q-network (RDQN) having a local attention mechanism located between a convolutional neural network (CNN) and a long-short time memory (LSTM), a plurality of feature maps are generated by the CNN from an input image. Hard-attention is applied by the local attention mechanism to the generated plurality of feature maps by selecting a subset of the generated feature maps. Soft attention is applied by the local attention mechanism to the selected subset of generated feature maps by providing weights to the selected subset of generated feature maps in obtaining weighted feature maps.Type: ApplicationFiled: January 25, 2022Publication date: July 14, 2022Inventors: Yiwen GUO, Yuqing Hou, Anbang Yao, Dongqi Cai, Lin Xu, Ping Hu, Shandong Wang, Wenhua Cheng, Yurong Chen, Libin Wang
-
Patent number: 11341368Abstract: Methods and systems for advanced and augmented training of deep neural networks (DNNs) using synthetic data and innovative generative networks. A method includes training a DNN using synthetic data, training a plurality of DNNs using context data, associating features of the DNNs trained using context data with features of the DNN trained with synthetic data, and generating an augmented DNN using the associated features.Type: GrantFiled: April 7, 2017Date of Patent: May 24, 2022Assignee: Intel CorporationInventors: Anbang Yao, Shandong Wang, Wenhua Cheng, Dongqi Cai, Libin Wang, Lin Xu, Ping Hu, Yiwen Guo, Liu Yang, Yuqing Hou, Zhou Su, Yurong Chen
-
Patent number: 11263490Abstract: Methods and systems for budgeted and simplified training of deep neural networks (DNNs) are disclosed. In one example, a trainer is to train a DNN using a plurality of training sub-images derived from a down-sampled training image. A tester is to test the trained DNN using a plurality of testing sub-images derived from a down-sampled testing image. In another example, in a recurrent deep Q-network (RDQN) having a local attention mechanism located between a convolutional neural network (CNN) and a long-short time memory (LSTM), a plurality of feature maps are generated by the CNN from an input image. Hard-attention is applied by the local attention mechanism to the generated plurality of feature maps by selecting a subset of the generated feature maps. Soft attention is applied by the local attention mechanism to the selected subset of generated feature maps by providing weights to the selected subset of generated feature maps in obtaining weighted feature maps.Type: GrantFiled: April 7, 2017Date of Patent: March 1, 2022Assignee: Intel CorporationInventors: Yiwen Guo, Yuqing Hou, Anbang Yao, Dongqi Cai, Lin Xu, Ping Hu, Shandong Wang, Wenhua Cheng, Yurong Chen, Libin Wang