Abstract: A digital signage system displays a plurality of sets of electronic content on a touch panel display so as to be selectable. The system sets a session for accepting the selection of content, accepts the selection of content by the user, and temporarily holds the information about the content selected by the user during the session. After the selection of content, the system creates access information for collectively accessing the held content, issues a two-dimensional code that is based on the access information and can be read using the user's mobile terminal, and displays the two-dimensional code on the touch panel display.

Abstract: Provided are approaches for customizing transaction cards using a multi-part card body. In some approaches, the transaction card may include a first section comprising a first outer perimeter, a first interior edge, and a second interior edge extending from the first interior edge, wherein the first and second interior edges form an obtuse angle, and wherein a slot is formed in the first interior edge. The transaction card may further include a second section couplable with the first section, wherein the second section comprises a second outer perimeter, a third interior edge, and a fourth interior edge extending from the third interior edge, wherein a protrusion extends from the third interior edge, and wherein the protrusion is retained within the slot when the first section and the second section are joined together. The transaction card may further include a chip coupled to the first section.

Abstract: An IC card according to an embodiment includes a contact communication interface, a non-contact communication interface, and a processor. A contact communication means is configured to perform contact communication with a reader/writer. A non-contact communication means is configured to perform non-contact communication with the reader/writer. The processor is configured to set an Answer To Reset (ATR) in the contact communication and an Answer To Select (ATS) in the non-contact communication based on a predetermined command including historical bytes.

Abstract: The disclosure provides an electronic cigarette with NFC anti-counterfeiting code and an anti-counterfeiting method thereof. The electronic cigarette comprises a vaporizer arranged with an insertion portion, and a battery rod arranged with a receiving portion for accommodating the insertion portion. The battery rod comprises a battery, and a control circuit board arranged with a microcontroller and a switching circuit. When the insertion portion is inserted in the receiving portion such that the NFC tag and the NFC tag antenna of the insertion portion are close to the NFC card reader and the NFC card reader antenna of the receiving portion, the NFC card reader automatically reads the anti-counterfeiting code preset in the NFC tag and sends it to the microcontroller for identification. If the anti-counterfeiting code is correct, the microcontroller MCU controls switching on of the switching circuit to allow the electronic cigarette to enter a standby state.

Abstract: The present application provides a device supporting card-swiping, which relates to the technical field of communications. The device supporting card-swiping comprises a display screen (1), a protective back plate (2) and a contactless card reader (3). Wherein the antenna (310) of the contactless card reader (3) is arranged between the display screen (1) and the protective back plate (2) for sensing a user's card swiping signal. The device supporting card-swiping can increase the effective card swiping distance and improve user experience.

Abstract: A device detecting system is provided. The device detecting system includes a bar code scanner, a plurality of device accommodating spaces, a screen, and a server. The server obtains bar code information via the bar code scanner and opens one of the device accommodating spaces based on the bar code information to accommodate an electronic device. The server performs a test procedure on the electronic device to generate a test result, and displays the test result and operation information corresponding to the test result on the screen.

Abstract: In an induction system for use with an object processing system, a distribution system is disclosed for providing dissimilar objects into one of a plurality of receiving units. The distribution system includes a conveyor section that travels over a vacuum roller, and the plurality of receiving units arranged generally below the vacuum roller such that an object with a mass will fall toward one of the plurality of receiving units based on both the object's mass and a vacuum force applied to the object through the conveyor section.

Abstract: At least some embodiments of the present invention are directed to barcode readers having weigh platters. In an example, a barcode reader includes a weigh platter having a central region, a proximal edge region, a first lateral edge region, a second lateral edge region, and a distal edge region, where a the first lateral edge of the first lateral edge region and a second lateral edge of the second lateral edge region extends above a transverse plane defined by the central region.

Abstract: A method, system and barcode reader for transmitting barcode scan information from a barcode reader to an external imaging system including one or more image capturing devices when the barcode reader is within a field of view of the external imaging system. Responsive to successfully decoding an indicia, the barcode reader transmits, to the external imaging system via an optical signal assembly, a first optical signal based at least in part on the indicia. Responsive to detecting an object within a product scanning region of the barcode reader and no successful decode of an indicia, the barcode reader transmits, to the external imaging system via the optical signal assembly, a second optical signal indicating the object being passed within the product scanning region with no successful decode.

Abstract: A tooling system may comprise a device configured to communicate with an identifier of a machine tool and a tool database having tool information including a predefined proximity associated with the identifier. The device may be configured to automatically activate, without a user-initiated scan, communication with at least one of the machine tool and the tool database in response to the identifier being within the predefined proximity. The device may also be configured to transfer tool information of the machine tool with respect to the tool database.

Abstract: The present description relates to a 2D visual code based on the superposition of pure flat waves framed in black, designed to be read at long distances, with different types of degradation (noise, blurring or movement) and in adverse lighting conditions. To this end, the information carrying elements are waves that are distributed over the whole image. The combination of waves with different directions, frequencies, amplitudes and phases permits generating a high number of configurations that may be decoded in a robust and efficient manner.

Abstract: A system and method for counting items are described. A photocell detects an item, after which a converter converts the generated digital signal to an audio signal that a computing device can recognize. Data from the counting of items is displayed on a user interface of the computing device and/or other visual interfaces.

Abstract: A comparative rejection sampling technique selects a bound set of possible execution run results for a process to be simulated, such as a quantum circuit, for each possible execution run result a modeled probability of a state associated with the possible execution run result is determined. For example a tensor network algorithm may be used to determine quantum state probabilities for each quantum state execution result included in the bound set. Based on the determined probabilities one of the possible execution run results is selected from the set of possible execution run results as an accepted simulated execution run result for a run of the process being simulated.

Abstract: A method of reducing stray coupling in a qubit array, includes turning ON a first coupler between a first qubit and second qubit of the qubit array by providing a pulse having a first amplitude to the first coupler. A stray coupling between the first coupler and a spectator qubit is reduced by turning ON a second coupler coupled to the spectator qubit, by providing a compensation pulse having a second amplitude, to the second coupler, based on the pulse having the first amplitude.

Abstract: Systems, methods, applications and uses for trans-radix quantum information processing elements are disclosed.

Abstract: A quantum controller comprises a quantum control pulse generation circuit and digital signal management circuit. The quantum control pulse generation circuit is operable to generate a quantum control pulse which can be processed by any of a plurality of controlled circuits, and generate a first digital signal which can be routed to any of the plurality of controlled circuits. The digital signal management circuit is operable to detect, during runtime, to which one or more of the plurality of controlled circuits the first digital signal is to be routed, to manipulate the first digital signal based on the one or more of the plurality of controlled circuits to which the first digital signal is to be routed, where the manipulation results in one or more manipulated digital signals, and to route the one or more manipulated digital signals to one or more of the plurality of controlled circuits.

Abstract: Preventing quantum errors using a Quantum Error Correction Algorithm Trainer (QECAT) table is disclosed herein. In one example, a processor device of a computing device is to identify a subset of a plurality of QECAT table entries of a QECAT table, wherein each QECAT table entry of the subset corresponds to an occurrence of a quantum error. The processor device is further to obtain metadata from each QECAT table entry of the subset. The processor device identifies, based on the metadata from each QECAT table entry of the subset, a common characteristic of each occurrence of the quantum error. The processor device then determines, based on the common characteristic, a preventative action to prevent a future occurrence of the quantum error.

Abstract: A method comprises receiving a network of a plurality of nodes and a plurality of edges, each of the nodes comprising members representative of at least one subset of training data points, each of the edges connecting nodes that share at least one data point, grouping the data points into a plurality of groups, each data point being a member of at least one group, creating a first transformation data set, the first transformation data set including the training data set as well as a plurality of feature subsets associated with at least one group, values of a particular data point for a particular feature subset for a particular group being based on values of the particular data point if the particular data point is a member of the particular group, and applying a machine learning model to the first transformation data set to generate a prediction model.

Abstract: A machine learning algorithm, such as a random forest regressor, can be trained using a set of annotated data objects to estimate the risk or business value for an object. The feature contributions for each data object can be analyzed and a representation generated that clusters data objects by feature contributions. Any clustering of data objects with incorrect scores in the visualization can be indicative of gaps in the regressor training. Adjustments to the inputs can be made, and the regressor retrained, to eliminate clustering of errors for similar feature contributions. Correcting the risk score estimations can ensure that the appropriate security policies and permissions are applied to each data object.

Abstract: An input unit receives an input of data, as learning purpose data and determination target data, in which requests made to a server by a user are represented in a time series. Then, a shaping unit shapes the received data. A classifying unit classifies the shaped data for each user who made the requests. Then, a learning unit extracts, from the classified learning purpose data, consecutive n requests as feature values of the learning purpose data, performs learning by using the feature values of the learning purpose data, and creates a profile for each user. A determination unit extracts, from the classified determination target data, consecutive n requests as feature values of the determination target data and performs determination of the determination target data based on the feature values of the determination target data and based on the profiles created by the learning unit.

Abstract: Herein are techniques that train regressor(s) to predict how effective would a machine learning model (MLM) be if trained with new hyperparameters and/or dataset. In an embodiment, for each training dataset, a computer derives, from the dataset, values for dataset metafeatures. The computer performs, for each hyperparameters configuration (HC) of a MLM, including landmark HCs: configuring the MLM based on the HC, training the MLM based on the dataset, and obtaining an empirical quality score that indicates how effective was said training the MLM when configured with the HC. A performance tuple is generated that contains: the HC, the values for the dataset metafeatures, the empirical quality score and, for each landmark configuration, the empirical quality score of the landmark configuration and/or the landmark configuration itself. Based on the performance tuples, a regressor is trained to predict an estimated quality score based on a given dataset and a given HC.

Abstract: In exemplary aspects, a golden data structure can be used to validate the stability of machine learning (ML) models and weights. The golden data structure includes golden input data and corresponding golden output data. The golden output data represents the known correct results that should be output by a ML model when it is run with the golden input data as inputs. The golden data structure can be stored in a secure memory and retrieved for validation separately or together with the deployment of the ML model for a requested ML operation. If the golden data structure is used to validate the model and/or weights concurrently with the performance of the requested operation, the golden input data is combined with the input data for the requested operation and run through the model. Relevant outputs are compared with the golden output data to validate the stability of the model and weights.

Abstract: A method comprises receiving a network of a plurality of nodes and a plurality of edges, each of the nodes of the plurality of nodes comprising members representative of at least one subset of initial data points, selecting a subset of the data points based on each node of the plurality of nodes, for each selected data point of the set of selected data points, determining a predetermined number of other data points that are closest in distance to that particular selected data point, grouping the selected data points into a plurality of groups based, at least in part, on the predetermined number of other data points of the set of selected data points that are closest in distance, each group of the plurality of groups including a different subset of data points, and providing a list of selected data points and the plurality of groups.

Abstract: A video augmentation apparatus is shown. The apparatus may comprise at least a processor and a memory. The processor may be configured to receive a plurality of videos, Additionally, the processor may generate a segment datum as a function of the plurality of videos. The segment datum may be classified to an augmentation datum. The classification may include training an augmentation classifier using a segment training data wherein the segment training data contains a plurality of data entries correlating required segment datum as an input to the augmentation datum as outputs. The classification may further include generating an augmentation classification datum, wherein augmentation classification datum is generated by classifying the segment datum to the augmentation datum using the augmentation classifier. The processor then may generate an augmented video as a function of the augmentation classification datum and display the augmented video using a user display device.

Abstract: Systems and techniques for distributed machine learning (DML) in an information centric network (ICN) are described herein. Finite message exchanges, such as those used in many DML exercises, may be efficiently implemented by treating certain data packets as interest packets to reduce overall network overhead when performing the finite message exchange. Further, network efficiency in DML may be improved achieved by using local coordinating nodes to manage devices participating in a distributed machine learning exercise. Additionally, modifying a round of DML training to accommodate available participant devices, such as by using a group quality of service metric to select the devices, or extending the round execution parameters to include additional devices, may have an impact on DML performance.

Abstract: Disclosed herein are systems and methods for determining data structures. In some embodiments, a classifier may be used to determine one or more attributes of an entity. In some embodiments, a clustering algorithm may be used to determine an attribute cluster. In some embodiments, an impact metric machine learning model may be used to determine an outlier cluster. In some embodiments, an outlier process may be determined as a function of the outlier cluster. In some embodiments, a visual element may be determined as a function of an outlier process and may be displayed to a user.

Abstract: Systems and methods may use one or more artificial intelligence models that predict an effect of a predicted event on a current state of the system. The systems and methods may use one or more artificial intelligence models that predict an effect and/or occurrence of a predicted event based on the current state of the system. In order to generate responses that are both timely and pertinent (e.g., in a dynamic fashion), the system must determine, both quickly (i.e., in real-time or near real-time) and accurately, the predicted event.

Abstract: A method includes receiving, by a server computer, data of a communication device; training, by the server computer, a neural network model based on the data of the communication device and communication device metadata from one or more additional communication devices, to generate a machine learning model configured to determine, based on a metadata associated with an application, a security value related to an indication of a security threat; and transmitting the machine learning model to the communication device. The communication device can use the machine learning model to determine the security value, by inputting the metadata associated with the application, as a vectorized data into the machine learning model. The communication device can determine whether to run or install the application based upon the security value.

Abstract: A method of modelling data, comprising: training an objective function of a linear classifier, based on a set of labeled data, to derive a set of classifier weights; defining a posterior probability distribution on the set of classifier weights of the linear classifier; approximating a marginalized loss function for an autoencoder as a Bregman divergence, based on the posterior probability distribution on the set of classifier weights learned from the linear classifier; and classifying unlabeled data using the autoencoder according to the marginalized loss function.

Abstract: A method for performing automatic visual inspection includes: capturing visual information of an object using a scanning system including a plurality of cameras; extracting, by a computing system including a processor and memory, one or more feature maps from the visual information using one or more feature extractors; classifying, by the computing system, the object by supplying the one or more feature maps to a complex classifier to compute a classification of the object, the complex classifier including: a plurality of simple classifiers, each simple classifier of the plurality of simple classifiers being configured to compute outputs representing a characteristic of the object; and one or more logical operators configured to combine the outputs of the simple classifiers to compute the classification of the object; and outputting, by the computing system, the classification of the object as a result of the automatic visual inspection.

Abstract: Methods, systems, and computer storage media for implementing neural networks in fixed point arithmetic computing systems. In one aspect, a method includes the actions of receiving a request to process a neural network using a processing system that performs neural network computations using fixed point arithmetic; for each node of each layer of the neural network, determining a respective scaling value for the node from the respective set of floating point weight values for the node; and converting each floating point weight value of the node into a corresponding fixed point weight value using the respective scaling value for the node to generate a set of fixed point weight values for the node; and providing the sets of fixed point floating point weight values for the nodes to the processing system for use in processing inputs using the neural network.

Abstract: A system includes receiving data associated with an account, the data having a plurality of members; generating based on an ensemble teacher model, a deep learning model having a number of layers; inputting a plurality of members determined to be daily inputs into the deep learning model; extracting a daily pattern from the daily inputs and aggregating a deep learning model output; inputting the global inputs and an aggregated deep learning model output into a classifier; outputting from the classifier, a number of scores combined into a single score for the account. Further, the device may include alerting a user if the single score falls outside of a predetermined threshold.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for training a neural network system used to control an agent interacting with an environment. One of the methods includes receiving a current observation; processing the current observation using a proposal neural network to generate a proposal output that defines a proposal probability distribution over a set of possible actions that can be performed by the agent to interact with the environment; sampling (i) one or more actions from the set of possible actions in accordance with the proposal probability distribution and (ii) one or more actions randomly from the set of possible actions; processing the current observation and each sampled action using a Q neural network to generate a Q value; and selecting an action using the Q values generated by the Q neural network.

Abstract: Described herein is a neural network accelerator (NNA) with a decompression unit that can be configured to perform multiple types of decompression. The decompression may include a separate subunit for each decompression type. The subunits can be coupled to form a pipeline in which partially decompressed results generated by one subunit are input for further decompression by another subunit. Depending on which types of compression were applied to incoming data, any number of the subunits may be used to produce a decompressed output. In some embodiments, the decompression unit is configured to decompress data that has been compressed using a zero value compression scheme, a shared value compression scheme, or both. The NNA can also include a compression unit implemented in a manner similar to that of the decompression unit.

Abstract: Disclosed is a method for implementing an adaptive stochastic spiking neuron based on a ferroelectric field effect transistor, relating to the technical field of spiking neurons in neuromorphic computing. Hardware in the method includes a ferroelectric field effect transistor (fefet), an n-type mosfet, and an l-fefet formed by enhancing a polarization degradation characteristic of a ferroelectric material for the ferroelectric field-effect transistor, wherein a series structure of the fefet and the n-type mosfet adaptively modulates a voltage pulse signal transmitted from a synapse. The l-fefet has a gate terminal connected to a source terminal of the fefet to receive the modulated pulse signal, and simulates integration, leakage, and stochastic spike firing characteristics of a biological neuron, thereby implementing an advanced function of adaptive stochastic spike firing of the neuron.

Abstract: The present invention relates to the field of smart grids, and provides a non-intrusive load monitoring method and device based on temporal attention mechanism. The method comprises the following steps: obtaining a total load data, an equipment load data, and corresponding sampling time of a building during a certain period of time; integrating the total load data and the equipment load data with the corresponding sampling time to obtain an enhanced total load data and an enhanced equipment load data; using a sliding window method to segment the enhanced total load data and the enhanced equipment load data, and constructing a deep learning training dataset; constructing a neural network model based on a deep learning training framework and training the model using the training dataset. The present invention can effectively extract the working time mode of the load and its inherent dependencies, thereby improving the accuracy of load monitoring.

Abstract: A neuromorphic apparatus configured to process a multi-bit neuromorphic operation including a single axon circuit, a single synaptic circuit, a single neuron circuit, and a controller. The single axon circuit is configured to receive, as a first input, an i-th bit of an n-bit axon. The single synaptic circuit is configured to store, as a second input, a j-th bit of an m-bit synaptic weight and output a synaptic operation value between the first input and the second input. The single neuron circuit is configured to obtain each bit value of a multi-bit neuromorphic operation result between the n-bit axon and the m-bit synaptic weight, based on the output synaptic operation value. The controller is configured to respectively determine the i-th bit and the j-th bit to be sequentially assigned for each time period of different time periods to the single axon circuit and the single synaptic circuit.

Abstract: Some embodiments provide a neural network inference circuit for executing a neural network that includes multiple nodes that use state data from previous executions of the neural network. The neural network inference circuit includes (i) a set of computation circuits configured to execute the nodes of the neural network and (ii) a set of memories configured to implement a set of one or more registers to store, while executing the neural network for a particular input, state data generated during at least two executions of the network for previous inputs. The state data is for use by the set of computation circuits when executing a set of the nodes of the neural network for the particular input.

Abstract: In one example, an apparatus comprises: a direct memory access (DMA) descriptor queue that stores DMA descriptors, each DMA descriptor including an indirect address; an address translation table that stores an address mapping between indirect addresses and physical addresses; and a DMA engine configured to: fetch a DMA descriptor from the DMA descriptor queue to the address translation table to translate a first indirect address of the DMA descriptor to a first physical address based on the address mapping, and perform a DMA operation based on executing the DMA descriptor to transfer data to or from the first physical address.

Abstract: Disclosed is a convolution operator system comprising a Convolution Neural Network (CNN) reconfigurable engine including a plurality of Mini Parallel Rolling Engines (MPREs) for performing a convolution operation concurrently on an image. An input router receives image data. A controller allocates image data to computing blocks through a set of data flow control blocks. Each computing block produces a convolution output corresponding to each row of the image. The controller allocates a plurality of group having one or more computing blocks to generate a set of convolution output. Further, a pipeline adder aggregates the set of convolution output to produce an aggregated convolution output. An output router transmits either the convolution output or the aggregated convolution output for performing subsequent convolution operation to generate a convolution result for the image data.

Abstract: A 2D array-based neuromorphic processor includes: axon circuits each being configured to receive a first input corresponding to one bit from among bits indicating n-bit activation; first direction lines extending in a first direction from the axon circuits; second direction lines intersecting the first direction lines; synapse circuits disposed at intersections of the first direction lines and the second direction lines, and each being configured to store a second input corresponding to one bit from among bits indicating an m-bit weight and to output operation values of the first input and the second input; and neuron circuits connected to the first or second direction lines, each of the neuron circuits being configured to receive an operation value output from at least one of the synapse circuits, based on time information assigned individually to the synapse circuits, and to perform an arithmetic operation by using the operation values.

Abstract: Provided are systems and methods for operating a neural network processor, wherein the processor includes an input selector circuit that can be configured to select the data that will be input into the processor's computational array. In various implementations, the selector circuit can determine, for a row of the array, whether the row input will be the output from a buffer memory or data that the input selector circuit has selected for a different row. The row can receive an input feature map from a set of input data or an input feature map that was selected for inputting into a different row, such that the input feature map is input into more than one row at a time. The selector circuit can also include a delay circuit, so that the duplicated input feature map can be input into the computational array later than the original input feature map.

Abstract: Disclosed is a neuromorphic integrated circuit including, in some embodiments, a multi-layered neural network disposed in an analog multiplier array of two-quadrant multipliers. Each multiplier of the multipliers is wired to ground and draws a negligible amount of current when input signal values for input signals to transistors of the multiplier are approximately zero, weight values of the transistors of the multiplier are approximately zero, or a combination thereof. Also disclosed is a method of the neuromorphic integrated circuit including, in some embodiments, training the neural network; tracking rates of change for the weight values; determining if and how quickly certain weight values are trending toward zero; and driving those weight values toward zero, thereby encouraging sparsity in the neural network. Sparsity in the neural network combined with the multipliers wired to ground minimizes power consumption of the neuromorphic integrated circuit such that battery power is sufficient for power.

Abstract: An arithmetic device and an electronic device having small power consumption is provided. An arithmetic device and an electronic device capable of high-speed operation is provided. An arithmetic device and an electronic device capable of suppressing heat generation is provided. The arithmetic device includes a first arithmetic portion and a second arithmetic portion. The first arithmetic portion includes a first CPU core and a second CPU core. The second arithmetic portion includes a first GPU core and a second GPU core. The CPU cores each have a power gating function and each include a first data retention circuit electrically connected to a flip-flop. The first GPU core includes a second data retention circuit capable of retaining an analog value and reading out the analog value as digital data of two or more bits. The second GPU core includes a third data retention circuit capable of retaining a digital value and reading out the digital value as digital data of one bit.

Abstract: Disclosed herein are techniques for implementing a large fully-connected layer in an artificial neural network. The large fully-connected layer is grouped into multiple fully-connected subnetworks. Each fully-connected subnetwork is configured to classify an object into an unknown class or a class in a subset of target classes. If the object is classified as the unknown class by a fully-connected subnetwork, a next fully-connected subnetwork may be used to further classify the object. In some embodiments, the fully-connected layer is grouped based on a ranking of target classes.

Abstract: Computer-implemented methods and systems are provided for automatically performing a task on a remote computer. During a registration stage, the system receives personal information of a human user, obtains an IP address and a device configuration for a computing device, and stores the personal information, IP address, and the device configuration in record in a database. The system receives a request to interact with a remote website to perform a task, the request including the personal information of the human user. The system then retrieves the record from the database using the personal information. The system creates a virtual machine based on the device configuration for the computing device, selects one of a geographically distributed set of proxy servers having an IP geographically address resembling the IP address for the computing device, and executes instructions causing the virtual machine to interact with the remote website using the proxy server to perform the task.

Abstract: An interconnect topology for communication between GPUs in a computing system is determined. A quantity of directed spanning trees are generated for transmitting data between the GPUs using the interconnect topology and packed. The directed spanning trees define the connections between GPUs that are to be utilized for the transmission and the amount of data to be transmitted on each connection. Program code is generated for implementing the data transfer defined by the directed spanning trees. When the program code is executed, the directed spanning trees are used to pipeline the transmission of chunks of data, such as model parameters used during data-parallel DNN training, between the GPUs. The program code can also determine an optimal chunk size for data to be transferred between the GPUs.

Abstract: An artificial intelligence laundry drying machine is provided. The artificial intelligence laundry drying machine includes a communication unit configured to receive washing information of a washing machine from the washing machine, a drying unit configured to dry a drying object, and a processor configured to acquire information about a dryness degree of the drying object by using the washing information and perform drying by using the information about the dryness degree.

Abstract: An off-policy reinforcement learning actor-critic neural network system configured to select actions from a continuous action space to be performed by an agent interacting with an environment to perform a task. An observation defines environment state data and reward data. The system has an actor neural network which learns a policy function mapping the state data to action data. A critic neural network learns an action-value (Q) function. A replay buffer stores tuples of the state data, the action data, the reward data and new state data. The replay buffer also includes demonstration transition data comprising a set of the tuples from a demonstration of the task within the environment. The neural network system is configured to train the actor neural network and the critic neural network off-policy using stored tuples from the replay buffer comprising tuples both from operation of the system and from the demonstration transition data.

Abstract: A system and method of detecting an aberrant message is provided. An ordered set of words within the message is detected. The set of words found within the message is linked to a corresponding set of expected words, the set of expected words having semantic attributes. A set of grammatical structures represented in the message is detected, based on the ordered set of words and the semantic attributes of the corresponding set of expected words. A cognitive noise vector comprising a quantitative measure of a deviation between grammatical structures represented in the message and an expected measure of grammatical structures for a message of the type is then determined. The cognitive noise vector may be processed by higher levels of the neural network and/or an external processor.