31 results about "Reservoir computing" patented technology

To realize a reservoir computing system with a small size and reduced learning cost, provided is a laser apparatus including a laser; a feedback waveguide that is operable to feed light output from the laser back to the laser; an optical splitter that is provided in a path of the feedback waveguide and is operable to output a portion of light propagated in the feedback waveguide to outside; and a first ring resonator that is operable to be optically connected to the feedback waveguide, as well as a reservoir computing system including this laser apparatus.

The invention discloses a new energy power plant station management platform, which comprises a host storage layer, a network communication layer, a support software layer, a data resource layer, an application system layer and a user interaction layer. The host storage layer provides computing, storage and disaster recovery resources. The network communication layer completes operating data access of a field new energy power plant station, and supports data transmission and exchange in wireless and wired manners. The support software layer is used for scheduling and expanding the computing and storage resources of the host storage layer by means of virtualization software. The data resource layer realizes persistent storage and use of various types of business subject data by adopting a traditional relational database and various NoSQL databases according to different data features. The application system layer transmits new energy power plant data to the management personnel in realtime, provides online control and plant station management, and utilizes data of the data resource layer to perform fault prediction and diagnosis by means of data pattern identification and adoptinga model training algorithm. The user interaction layer is used for establishing user interaction channels for different application scenarios.

To realize a reservoir computing system easily implemented as hardware, provided is a reservoir computing system including a reservoir operable to output an inherent output signal in response to an input signal. An input node is operable to supply the reservoir with an input signal corresponding to input data, and an output node is operable to output an output value corresponding to an output signal that is output by the reservoir in response to the input data. An adaptive filter is operable to output output data based on a result obtained by weighting a plurality of the output values output from the output node at a plurality of timings with a plurality of weights. Also provided are a learning method and a computer program product.

Described is a cognitive signal processor that can denoise an input signal that contains a mixture of waveforms over a large bandwidth. Delay-embedded mixture signals are generated from a mixture of input signals. The delay-embedded mixture signals are mapped with a reservoir computer to reservoir states of a dynamical reservoir having output layer weights. The output layer weights are adapted based on short-time linear prediction. Finally, a denoised output of the mixture of input signals is generated.

Described is a neuromorphic processor for signal denoising and separation. The neuromorphic processor generates delay-embedded mixture signals from an input mixture of pulses. Using a reservoir computer, the delay-embedded mixture signals are mapped to reservoir states of a dynamical reservoir having output layer weights. The output layer weights are adapted based on short-time linear prediction, and a denoised output of the mixture of input signals us generated. The denoised output is filtered through a set of adaptable finite impulse response (FIR) filters to extract a set of separated narrowband pulses.

An artificial neuron includes a signal mixer that combines input signals to provide a first stochastic bit-stream as output and a stochastic activation function circuit configured to receive the first stochastic bit-stream from the signal mixer and to generate therefrom a second stochastic bit-stream. The first stochastic bit-stream is representative of a first output value. In the stochastic activation function circuit, n independent stochastic bit-streams, each representative of the first output value, are summed to provide a selection signal that is provided to a multiplexer to select between n+1 coefficient bit-streams and provide the second stochastic bit-stream. The activation function has a characteristic determined by the proportion of ones in each of the n+1 coefficients bit-streams. One or more artificial neurons may be used in an Artificial Neural Network, such as a Time Delay Reservoir network.

The present invention discloses a reservoir connectivity analysis method based on multi-target analysis. The method comprises the following steps of S1 determining reservoirs and the basic situations of the reservoirs which are involved in a connecting engineering needing to be demonstrated; S2 designing corresponding connecting schemes according to the conditions, such as the reservoir distribution situation, the reservoir water supply tasks, etc.; S3 applying a multi-target decision method to establish a reservoir multi-target joint scheduling model; S4 adopting a multi-target genetic algorithm epsilon-NSGAII to optimize and obtain the multi-target tradeoff solution sets of the reservoir scheduling under the schemes; S5 utilizing the visualized analysis to compare the multi-target solution sets of the schemes, and evaluating the advantages and disadvantages of the multi-target solution sets of the schemes, thereby giving out a reservoir connecting feasibility analysis result. The reservoir connectivity analysis method based on multi-target analysis of the present invention fully considers a watershed hydrology compensation effect and an inter-reservoir storage capacity compensation effect, has stronger persuasion for the feasibility analysis of the connecting engineering, and facilitates guiding the management decision personnel to make the reasonable decisions about the reservoir connectivity problem.

To realize a reservoir computing system easily implemented as hardware, provided is a reservoir computing system including a reservoir operable to output an inherent output signal in response to an input signal. An input node is operable to supply the reservoir with an input signal corresponding to input data, and an output node is operable to output an output value corresponding to an output signal that is output by the reservoir in response to the input data. An adaptive filter is operable to output output data based on a result obtained by weighting a plurality of the output values output from the output node at a plurality of timings with a plurality of weights. Also provided are a learning method and a computer program product.

Described is a system for signal denoising using a cognitive signal processor having a time-varying reservoir. The system receives a noisy input signal of a time-series of data points from a mixture of waveform signals. The noisy input signal is linearly mapped into the time-varying reservoir. A high-dimensional state-space representation of the mixture of waveform signals is generated by combining the noisy input signal with a plurality of reservoir states. The system then generates a denoised signal corresponding to the noisy input signal.

The invention discloses a reservoir computing hardware implementation method and device based on a coupled MEMS resonator, and the method comprises the steps: carrying out the preprocessing of a to-be-detected time sequence signal, so as to enable the to-be-detected time sequence signal to correspond to a virtual node of the coupled MEMS resonator; designing a nonlinear vibration equation of the coupled MEMS resonator, and regulating and controlling the coupled MEMS resonator to a preset nonlinear working point according to the equation; respectively detecting two signal test ends of the MEMScoupled resonator to obtain a first output signal and a second output signal corresponding to the to-be-tested signal corresponding to each moment, and feeding back the output signal corresponding tothe to-be-tested signal at the current moment to the virtual node corresponding to the next moment through a bidirectional time delay feedback loop; and performing regression training on the preset target value and the first output signal and the second output signal corresponding to the to-be-measured signal corresponding to each moment to obtain a weight coefficient required by calculation of the storage pool. According to the method, the data mapping dimension and the memory performance are enhanced, and richer nonlinear characteristics are provided for reserve pool calculation.

A technology that can enhance the computing performance of a computing system using reservoir computing (RC), includes a computing system which performs computation using a recurrent neural network (RNN) including an input unit, a reservoir unit, and an output unit. The reservoir unit includes a plurality of nodes circularly connected to each other. The circular connection has a weight matrix for determining a weight between the nodes of the plurality of nodes, in which a weight between the nodes closely arranged on the circle is larger than a weight between the nodes arranged away from each other on the circle. The plurality of nodes each have a g value that is a parameter for designating nonlinearity of an activation function of each of the nodes, and that is set so as to periodically change in a direction on the circle.

A quantum state classifier comprises: a reservoir computing circuit for post-processing a quantum bit to obtain a readout signal; and a readout circuit, coupled to the reservoir computing circuit, for discriminating a quantum state of a qubit from the readout signal from among a plurality of possible quantum states. The readout circuitry is trained in a calibration process activated by a particular quantum state of each of the plurality of quantum states, respectively, such that the weights within the linear readout circuitry are updated by a small batch learning of each of the plurality of measurement sequences for the calibration process. The readout circuit generates a binary output after the plurality of measurement sequences during a post-calibration classification process for the test qubits. The quantum state classifier also includes a controller coupled to the readout circuitry, selectively actuatable to output a control pulse in response to a quantum state of the test qubit indicated by the binary output.

A reservoir computing data flow processor includes a plurality of reservoir units to be units constituting a reservoir. The reservoir is able to be reconfigured by changing a connection relationship between the reservoir units. Each of the reservoir units is an operation unit block configured to execute a predetermined operation. The operation unit block includes a first adder configured to perform an addition operation on at least two inputs, a nonlinear operator configured to apply a nonlinear function to an output from the first adder or a result of multiplying the output by a predetermined coefficient, and a second adder configured to perform an addition operation on at least two inputs including an output from the nonlinear operator or a result of multiplying the output by a predetermined coefficient.