Abstract: In one aspect, a method of training a DNN includes: providing a weight matrix (W) as a linear combination of matrices/arrays A and C; in a forward cycle, transmitting an input vector x through arrays A and C and reading output vector y; in a backward cycle, transmitting an error signal S through arrays A and C and reading output vector z; updating array A by transmitting input vector x and error signal S through array A; in a forward cycle, transmitting an input vector ei through array A and reading output vector y?; calculating ƒ(y?) using y?; and updating array C by transmitting input vector ei and ƒ(y?) through array C. A DNN is also provided.

Abstract: A computer-implemented method and computer processing system are provided for update management for a neural network. The method includes performing an isotropic update process on the neural network using a Resistive Processing Unit. The isotropic update process uses a multiplicand and a multiplier from a multiplication operation. The performing step includes scaling the multiplicand and the multiplier to have a same order of magnitude.

Abstract: A technique relates a resistive processing unit (RPU) array. A set of conductive column wires are configured to form cross-points at intersections between the set of conductive row wires and a set of conductive column wires. Two-terminal RPUs are hysteretic such that the two-terminal RPUs each have a conductance state defined by hysteresis, where a two-terminal RPU of the two-terminal RPUs is located at each of the cross-points.

Abstract: An artificial neural network and methods for performing computations on an artificial neural network include multiple neurons, including a layer of input neurons, one or more layers of hidden neurons, and a layer of output neurons. Arrays of weights are configured to accept voltage pulses from a first layer of neurons and to output current to a second layer of neurons during a feed forward operation. Each array of weights includes multiple resistive processing units having respective settable resistances.

Abstract: Technical solutions are described for implementing a convolutional neural network (CNN) using resistive processing unit (RPU) array. An example method includes configuring an RPU array corresponding to a convolution layer in the CNN based on convolution kernels of the layer. The method further includes performing forward pass computations via the RPU array by transmitting voltage pulses corresponding to input data to the RPU array, and storing values corresponding to output currents from the RPU arrays as output maps. The method further includes performing backward pass computations via the RPU array by transmitting voltage pulses corresponding to error of the output maps, and storing the output currents from the RPU arrays as backward error maps. The method further includes performing update pass computations via the RPU array by transmitting voltage pulses corresponding to the input data of the convolution layer and the error of the output maps to the RPU array.

Abstract: A system, comprising: a memory that stores computer-executable components; a processor, operably coupled to the memory, that executes the computer-executable components stored in the memory, wherein the computer-executable components comprise: an expression component that expresses the read current range in an RPU as read current Iwmin and Iwmax, a constant current source component that generates a reference current I, a computing component that subtracts the reference current value within from the read current value to generate an active net current read value that is negative, positive or null; a weighting component that analyzes the active current value and assigns it to a negative, positive or null weight.

Abstract: A system, comprising: a memory that stores computer-executable components; a processor, operably coupled to the memory, that executes the computer-executable components stored in the memory, wherein the computer-executable components comprise: an expression component that expresses the read current range in an RPU as read current Iwmin and Iwmax, a constant current source component that generates a reference current I, a computing component that subtracts the reference current value within from the read current value to generate an active net current read value that is negative, positive or null; a weighting component that analyzes the active current value and assigns it to a negative, positive or null weight.

Abstract: A method for monitoring at least one simulation program includes capturing, by a computer, a plurality of simulation data from the at least one simulation program, the capturing is performed in real time while the at least one simulation program is continuously streaming the plurality of simulation data, analyzing, by the computer, the captured plurality of simulation data using a streaming data software, identifying a plurality of predefined criteria within the analyzed plurality of simulation data, the plurality of predefined criteria includes at least one of an event, a result and a variable, and providing feedback to the at least one simulation program to modify a plurality of simulation parameters according to the at least one identified event, result and variable.

Abstract: An array of resistive processing units (RPUs) comprises a plurality of rows of RPUs and a plurality of columns of RPUs wherein each RPU comprises an AND gate configured to perform an AND operation of a first stochastic bit stream received from a first stochastic translator translating a number encoded from a neuron in a row and a second stochastic bit stream received from a second stochastic translator translating a number encoded from a neuron in a column. A first storage is configured to store a weight value of the RPU, and a second storage is configured to store an amount of change to the weight value of the RPU. When the first stochastic bit stream and the second stochastic bit stream coincide, the amount of change to the weight value of the RPU is added to the weight value of the RPU.

Abstract: A method for multiple copies of a set of multi-kernel set operations in a hardware accelerated neural network includes a word line for receiving a pixel value of an input image. A bit line communicates a modified pixel value. An analog memory cell including a first capacitor stores a first kernel weight of a first kernel in one of a plurality of kernel sets such that the pixel value is operated on by the first kernel weight to produce the modified pixel value. A charge connection connects the first capacitor to at least a second capacitor storing a second kernel weight of a related kernel of a second one of the plurality of kernel sets such that charge is shared between the first capacitor and at least the second capacitor to normalize the first kernel weight and the second kernel weight.

Abstract: Zero-shifting techniques in analog crosspoint arrays are provided. In one aspect, an analog array-based vector-matrix multiplication includes: a weight array connected to a reference array, each including a crossbar array having a set of conductive row wires and a set of conductive column wires intersecting the set of conductive row wires, and optimizable crosspoint devices at intersections of the set of conductive column wires and the set of conductive row wires. A method for analog array-based vector-matrix computing is also provided that includes: applying repeated voltage pulses to the crosspoint devices in the weight array until all of the crosspoint devices in the weight array converge to their own symmetry point; and copying conductance values for each crosspoint device from the weight array to the reference array.

Abstract: In some aspects, a method for performing analog matrix inversion on a matrix with a network of resistive device arrays B, W, Q, and C is described. The method may include initializing arrays W, Q, B and C, updating the connections of array W in parallel and array Q in parallel until a predetermined condition is satisfied, and responsive to determining that the predetermined condition is satisfied, outputting an inverted matrix based on outputs from the connections of arrays B, W, Q, and C.

Abstract: A photovoltaic device includes a first contact and a hybrid absorber layer. The hybrid absorber layer includes a chalcogenide layer and a semiconductor layer in contact with the chalcogenide layer. A buffer layer is formed on the absorber layer, and a transparent conductive contact layer is formed on the buffer layer.

Abstract: A photovoltaic device includes a first contact and a hybrid absorber layer. The hybrid absorber layer includes a chalcogenide layer and a semiconductor layer in contact with the chalcogenide layer. A buffer layer is formed on the absorber layer, and a transparent conductive contact layer is formed on the buffer layer.

Abstract: Mechanisms are provided for acceleration of convolutional neural networks on analog arrays. Input ports receive image signals from frames in an input image. Input memory arrays store the image signals received from the input ports into a respective input memory location to create a plurality of image sub-regions in input memory arrays. A distributor associated each of a set of analog array tiles in an analog array to a part of image sub-regions of the input memory arrays, so that one or more of a set of analog memory components is associated with the image signals in a distribution order to create a respective output signal. An assembler stores each of the respective output signals into one of a set of memory outputs in an output order that is determined by the distribution order.

Abstract: A resistive processing unit (RPU) device includes a weight storage device to store a weight voltage which corresponds to a weight value of the RPU device, and a read transistor having a gate connected to the weight storage device, and first and second source/drain terminals connected to first and second control ports, respectively. A current source connected to the second source/drain terminal generates a fixed reference current. The read transistor generates a weight current in response to the weight voltage. A read current output from the second control port represents a signed weight value of the RPU device. A magnitude of the read current is equal to a difference between the weight current and the fixed reference current. The sign of the read current is positive when the weight current is greater than the fixed reference current, and negative when the weight current is less than the fixed reference current.

Abstract: Embodiments of the present invention include a crossbar array that includes a resistive processing unit (RPU) device at each crosspoint in the crossbar array. The RPU device includes a single weight storage element, and multiple weight reader elements. A first weight reader element is coupled with a first row wire to compute a first matrix product value using a first value and a stored value, the first value being transmitted via the first row wire and the stored value being stored in the single weight storage element. A second weight reader element is coupled with a second row wire to compute a second matrix product value of a second value and said stored value, the second value being transmitted via the second row wire.

Abstract: A resistive processing unit (RPU) device includes a weight storage device to store a weight voltage which corresponds to a weight value of the RPU device, and a read transistor having a gate connected to the weight storage device, and first and second source/drain terminals connected to first and second control ports, respectively. A current source connected to the second source/drain terminal generates a fixed reference current. The read transistor generates a weight current in response to the weight voltage. A read current output from the second control port represents a signed weight value of the RPU device. A magnitude of the read current is equal to a difference between the weight current and the fixed reference current. The sign of the read current is positive when the weight current is greater than the fixed reference current, and negative when the weight current is less than the fixed reference current.

Abstract: Systems and methods are provided to perform weight update operations in a resistive processing unit (RPU) system to update weight values of RPU devices comprising tunable resistive device. A weight update operation for a given RPU device includes maintaining a weight update accumulation value for the RPU device, adjusting the weight update accumulation value by one unit update value in response to a detected coincidence of stochastic bits streams of input vectors applied on an update row and update column control lines connected to the RPU device, generating a weight update control signal in response to the accumulated weight value reaching a predefined threshold value, and adjusting a conductance level of the tunable resistive device by one unit conductance value in response to the weight update control signal, wherein the one unit conductance value corresponds to one unit weight value of the RPU device.

Abstract: A resistive processing unit (RPU) device includes a weight storage device to store a weight voltage which corresponds to a weight value of the RPU device, and a read transistor having a gate connected to the weight storage device, and first and second source/drain terminals connected to first and second control ports, respectively. A current source connected to the second source/drain terminal generates a fixed reference current. The read transistor generates a weight current in response to the weight voltage. A read current output from the second control port represents a signed weight value of the RPU device. A magnitude of the read current is equal to a difference between the weight current and the fixed reference current. The sign of the read current is positive when the weight current is greater than the fixed reference current, and negative when the weight current is less than the fixed reference current.