In a method of training a DNN, a weight matrix (W) is provided as a linear combination of matrices/arrays A and C. In a forward cycle, an input vector x is transmitted through arrays A and C and output vector y is read. In a backward cycle, an error signal δ is transmitted through arrays A and C and output vector z is read. Array A is updated by transmitting input vector x and error signal δ through array A. In a forward cycle, an input vector e.sub.i is transmitted through array A and output vector y′ is read. ƒ(y′) is calculated using y′. Array C is updated by transmitting input vector e.sub.i and ƒ(y′) through array C. A DNN is also provided.

This disclosure relates to method and system for providing personalized driving or navigation assistance. The method may include receiving sensory data with respect to a vehicle from a plurality of sensors and multi-channel input data with respect to one or more passengers inside the vehicle from a plurality of onboard monitoring devices, performing fusion of the sensory data and the multi-channel input data to generate multimodal fusion data, determining one or more contextual events based on the multi-modal fusion data using a machine learning model, wherein the machine learning model is trained using an incremental learning process and comprises a supervised machine learning model and an unsupervised machine learning model, analysing the one or more contextual events to generate a personalized driving recommendation, and providing the personalized driving recommendation to a driver passenger or a navigation device.

Systems, methods and apparatuses of processing overwhelming stimuli in vehicle data recorders. For example, a data recorder can have resources, such as memory components, a controller, an inference engine, etc. The resources can be partitioned into a first subset and a second subset. Abnormal stimuli in an input stream to the recorder may cause delay for real time processing. In response, a time sliced segment of the input stream is selected and assigned to the first subset; and a remaining segment is assigned to the second subset. The first and second subsets can separately process the time sliced segment and the remaining segment in parallel and thus avoid delay in the processing of the remaining segment. An artificial neural network (ANN) can determine a width for selecting the segment processed by the first subset; and the processing result can include a preferred width used to train the ANN.

A computer-implemented method is provided for machine prediction. The method includes forming, by a hardware processor, a Convolutional Dynamic Boltzmann Machine (C-DyBM) by extending a non-convolutional DyBM with a convolutional operation. The method further includes generating, by the hardware processor using the convolution operation of the C-DyBM, a prediction of a future event at time t from a past patch of time-series of observations. The method also includes performing, by the hardware processor, a physical action responsive to the prediction of the future event at time t.

A computer-implemented method is provided for machine prediction. The method includes forming, by a hardware processor, a Convolutional Dynamic Boltzmann Machine (C-DyBM) by extending a non-convolutional DyBM with a convolutional operation. The method further includes generating, by the hardware processor using the convolution operation of the C-DyBM, a prediction of a future event at time t from a past patch of time-series of observations. The method also includes performing, by the hardware processor, a physical action responsive to the prediction of the future event at time t.

Embodiments of the present invention provides a system and method of learning and classifying features to identify objects in images using a temporally coded deep spiking neural network, a classifying method by using a reconfigurable spiking neural network device or software comprising configuration logic, a plurality of reconfigurable spiking neurons and a second plurality of synapses. The spiking neural network device or software further comprises a plurality of user-selectable convolution and pooling engines. Each fully connected and convolution engine is capable of learning features, thus producing a plurality of feature map layers corresponding to a plurality of regions respectively, each of the convolution engines being used for obtaining a response of a neuron in the corresponding region. The neurons are modeled as Integrate and Fire neurons with a non-linear time constant, forming individual integrating threshold units with a spike output, eliminating the need for multiplication and addition of floating-point numbers.

Embodiments of the present invention provides a system and method of learning and classifying features to identify objects in images using a temporally coded deep spiking neural network, a classifying method by using a reconfigurable spiking neural network device or software comprising configuration logic, a plurality of reconfigurable spiking neurons and a second plurality of synapses. The spiking neural network device or software further comprises a plurality of user-selectable convolution and pooling engines. Each fully connected and convolution engine is capable of learning features, thus producing a plurality of feature map layers corresponding to a plurality of regions respectively, each of the convolution engines being used for obtaining a response of a neuron in the corresponding region. The neurons are modeled as Integrate and Fire neurons with a non-linear time constant, forming individual integrating threshold units with a spike output, eliminating the need for multiplication and addition of floating-point numbers.

Systems, methods and apparatus of recordation of vehicle data associated with errant vehicle behavior. For example, a vehicle includes: sensors configured to generate sensor data; control elements configured to generate control signals to be applied to the vehicle in response to user interactions with the control elements; electronic control units configured to provide status data in operations of the electronic control units; and a data storage device. The data storage device is configured to receive input data including the sensor data, the control signals and the status data, store the input data in a cyclic way in an input partition over time, generate a classification of errant behavior based on the input data and using an artificial neural network, and preserve a portion of the input data associated with the classification of errant behavior.

Systems, methods and apparatuses to detect an item left in a vehicle and to generate an alert about the item. For example, a camera configured in a vehicle can be used to monitor an item associated with a user of the vehicle. The item as in an image from the camera can be identified and recognized using an artificial neural network. In response to a determination that the item recognized in the image is left in the vehicle after the user has exited the vehicle, an alert is generated to indicate that an item is in the vehicle but the user is leaving the vehicle.

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for constructing and operating a recurrent artificial neural network. In one aspect, a method is for constructing nodes of an artificial recurrent neural network that mimics a target brain tissue. The method includes setting a total number of nodes in the artificial recurrent neural network, setting a number of classes and sub-classes of the nodes in the artificial recurrent neural network, setting structural properties of nodes in each class and sub-class, wherein the structural properties determine temporal and spatial integration of computation as a function of time as the node combines inputs, setting functional properties of nodes in each class and sub-class, wherein the functional properties determine activation, integration, and response functions as a function of time, setting a number of nodes in each class and sub-class of nodes, setting a level of structural diversity of each node in each class and sub-class of nodes and a level of functional diversity of each node in each class and sub-class of nodes, setting an orientation of each node, and setting a spatial arrangement of each node in the artificial recurrent neural network, wherein the spatial arrangement determines which nodes are in communication in the artificial recurrent neural network.