Intel Corporation

Yiwen Guo

Anbang Yao

Yurong Chen

Lin Xu

Shandong Wang

Ping Hu

Libin Wang

Dongqi Cai

Wenhua Cheng

Yuqing Hou

Ming Lu

Hao Zhao

Liu Yang

Zhou Su

Yan Li

Ignacio Javier Alvarez

Daniel Pohl

Ralf Graefe

Cornelius Buerkle

Xiangbin Wu

Fabian Oboril

Florian Geissler

Maik Sven Fox

Tao Kong

Yan Lin

Alexander Carl Unnervik

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. The weighted feature maps are stored in the LSTM. A Q value is calculated for different actions based on the weighted feature maps stored in the LSTM.

Methods and systems for budgeted and simplified training of deep neural networks

Techniques related to compressing a pre-trained dense deep neural network to a sparsely connected deep neural network for efficient implementation are discussed. Such techniques may include iteratively pruning and splicing available connections between adjacent layers of the deep neural network and updating weights corresponding to both currently disconnected and currently connected connections between the adjacent layers.

Dynamic neural network surgery

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.

Methods and apparatus for discriminative semantic transfer and physics-inspired optimization of features in deep learning

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.

Importance-aware model pruning and re-training for efficient convolutional neural networks

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.

Method and apparatus for reducing the parameter density of a deep neural network (DNN)

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.

Sensor calibration and sensor calibration detection

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. The second weak network is boosted using the determined classification error of the first weak network with adjusted weights. A second subset of training samples is processed by the second weak network using the adjusted weights.

Methods and systems for boosting deep neural networks for deep learning

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.

Methods and apparatus for enhancing a binary weight neural network using a dependency tree

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.

Methods and apparatus for enhancing a neural network using binary tensor and scale factor pairs

Described herein are hardware acceleration of random number generation for machine learning and deep learning applications. An apparatus () includes a uniform random number generator (URNG) circuit () to generate uniform random numbers and an adder circuit () that is coupled to the URNG circuit (). The adder circuit hardware () accelerates generation of Gaussian random numbers for machine learning.

Systems and methods for generating gaussian random numbers with hardware acceleration

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.

Methods and systems using camera devices for deep channel and convolutional neural network images and formats

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+layer. The nodes of each L+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+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.

Methods and systems using improved training and learning for deep neural networks

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.

Growing community of inventors

Beijing, China

Yiwen Guo