Abstract: A method of operating a procedural language and content generation system that involves correlating environment objects and object movement to input controls through operation of a correlator, operating an interpreter to evaluate the correlation of the input controls and object/object movement against known libraries to generate programmatic instructions, storing the programmatic instructions as an instruction set, transforming the instruction set into executable commands through a compiler, and configuring control logic to perform the executable commands in response to receiving detected environment objects and detected object movement from an image processor.

Abstract: Methods, computer program products, and systems are presented. The method computer program products, and systems can include, for instance: obtaining from a user one or more data queries; identifying a product of interest in response to the one or more data query; examining a plurality of product records to determine a set of related products that are related to the product of interest, wherein the examining includes performing image analysis to extract one or more product topic classifier from product image data representing one or more product; and providing one or more output in response to the examining.

Abstract: Concepts and technologies disclosed herein are directed to image prediction. According to one aspect disclosed herein, an image prediction system can receive a training data set that includes a plurality of training images. The image prediction system can define N-dimensional feature vectors corresponding to the plurality of training images in the training data set, parameterize the N-dimensional feature vectors to obtain a plurality of parameterized curves corresponding the plurality of training images in the training data set, obtain a square root velocity representation for each parameterized curve of the plurality of parameterized curves, rescale the plurality of parameterized curves to remove scaling variability among the plurality of parameterized curves, define a pre-shape space for the plurality of parameterized curves, and obtain shape space points pertaining to each parameterized curve of the plurality of parameterized curves on a shape space that inherits a structure from the pre-shape space.

Abstract: The present disclosure provides a method for joint manifold learning based heterogenous sensor data fusion, comprising: obtaining learning heterogeneous sensor data from a plurality sensors to form a joint manifold, wherein the plurality sensors include different types of sensors that detect different characteristics of targeting objects; performing, using a hardware processor, a plurality of manifold learning algorithms to process the joint manifold to obtain raw manifold learning results, wherein a dimension of the manifold learning results is less than a dimension of the joint manifold; processing the raw manifold learning results to obtain intrinsic parameters of the targeting objects; evaluating the multiple manifold learning algorithms based on the raw manifold learning results and the intrinsic parameters to determine one or more optimum manifold learning algorithms; and applying the one or more optimum manifold learning algorithms to fuse heterogeneous sensor data generated by the plurality sensors.

Abstract: Methods and systems are provided for identifying parameter image adjustments. In embodiments, a set of candidate parameter values associated with a parameter to be analyzed in association with an image is identified. Subsequently, the image is rendered in accordance with each candidate parameter value to generate a set of rendered images. A neural network can then be used to identify a parameter image adjustment to apply to the image based on features associated with the set of rendered images. The neural network can be trained based on a comparison of the identified parameter image adjustment and a reference parameter value associated with the parameter being analyzed.

Abstract: A system and method for pruning. A neural network includes a plurality of long short-term memory cells, each of which includes an input having a weight matrix Wc, an input gate having a weight matrix Wi, a forget gate having a weight matrix Wf, and an output gate having a weight matrix Wo. In some embodiments, after initial training, one or more of the weight matrices Wi, Wf, and Wo are pruned, and the weight matrix Wc is left unchanged. The neural network is then retrained, the pruned weights being constrained to remain zero during retraining.

Abstract: The present invention discloses a method operable on a digital electronic device comprising constrained processing unit employing a limited computer-readable storage medium, also known as a digital memory, for allowing an object classification process to be executed on an image. The object classification process may be allowed by a processing unit interlocking with a digital memory unit, which receives an image representing a digital image captured by light incident on an image sensor, denoted herein as an original image. In some cases, a computerized process operable on the digital processing may identify a list of pixel arrays located at the original image, and thereby allow a classification process to be operated on these pixel arrays. In some cases, a process operable on the digital processing may grant access to another computerized process to perform the classification process.

Abstract: A license plate reader uses a neural network to determine a plurality of predicted license plate regions within an image. The plurality of predicted license plate regions is transferred to an optical character recognition unit that performs optical character recognition on the plurality of predicted license plate regions to output a plurality of predicted character sequences. The license plate reader receives the output of the optical character recognition unit that contains the plurality of predicted character sequences and analyzes the plurality of predicted character sequences to determine a best estimate for a character sequence in the image.

Abstract: A request including data to be recorded in an information code is transmitted to a server from a terminal. Code generation information for generating an information code, in which the data is to be recorded, is received from the server. The information code is generated based on the received code generation information by the terminal. Practically, in response to the request, the server obtains an array pattern of the light-color and dark-color cells in the code region necessary for generating the information code in which the data is recorded. Information on the array pattern is set as code generation information and transmitted to the terminal. By another example, in the server, code print information including an array pattern of the light-color and dark-color cells in the code region can be obtained, and transmitted to the server for printing by the terminal.

Abstract: A transaction card may power on the transaction card using electric current induced from an interaction of the transaction card with an electromagnetic field. The transaction card may establish a communication with a device. The communication may indicate that the transaction card has powered. The transaction card may receive, from the device, a set of instructions to configure a set of applets on the transaction card after notifying the device that the transaction card has powered on. The set of applets to be configured may be related to completing one or more different transactions. The set of applets to be configured may be different than another set of applets already configured on the transaction card. The transaction card may configure the set of applets on the transaction card according to the set of instructions after receiving the set of instructions.

Abstract: Disclosed is an electronic device including: a body; a module enclosed in the body; a microcircuit and a direct current source, both forming part of the body; and an electronic component forming part of the module and accessible from the exterior, the electronic component being electrically connected to the microcircuit and having to be supplied with direct current from the direct current source. The body includes a first antenna connected to the direct current source via an oscillator and the module includes a second antenna connected to the electronic component via a rectifier circuit, the first and second antennas being electromagnetically coupled, providing a wireless radio frequency power supply to the electronic component of the module from the direct current source of the body.

Abstract: The present invention relates to the particularly innovative field of electronics applied to textiles. In particular, the present invention provides a method and a multilayer device which allows the use of known and present electronic passive electronic devices on the market or other electronic devices, making them one-piece with textiles or polymers through the use of heat-sealing materials and creating a stable innovative device.

Abstract: The present disclosure relates to the field of communications and, particularly to a visual card and an operating method for the visual card; the visual card includes a main control module, an antenna module, a power management module, a displaying module and a signal converting module; the main control module is connected to all of the antenna module, the power management module, the displaying module and the signal converting module, the signal converting module is connected to the antenna module; the main control module is configured to communicate with an external card read device via the antenna module, update and store key information; the main control module is further configured to monitor an exit signal according to an electronic signal input by the signal converting module.

Abstract: A wearable article, system, and method includes a structure configured to enclose a body part, a first antenna, in a first position on or within the structure, tuned to communicate according to a wireless communication modality through air, a second antenna, in a second position on or within the structure, tuned to communicate according to the wireless communication modality through the body part, the first antenna being tuned differently than the second antenna, and a transceiver, operatively coupled to at least one of the first antenna and the second antenna, configured to communicate with an external antenna via the at least one of the first and second antennas according to the wireless communication modality.

Abstract: A UHF band RFID tag with stable communication characteristics when attached to a metal surface. The UHF band RFID tag includes a resin block having a first surface serving as an attaching surface to the metal surface and a second surface opposing the first surface, a substrate on a second surface side of the resin block, a RFIC element mounted on the substrate, and a loop electrode connected to the RFIC element. The loop electrode includes metal bodies provided in the resin block and extending in a direction intersecting the first and second surfaces, a flat plate electrode on a first surface side of the resin block and connecting first ends of the metal bodies, conductor patterns on the second surface side of the resin block and connecting the RFIC element and second ends of metal bodies.

Abstract: An apparatus of operating a neural network is configured to compress one or more of activations or weights in one or more layer of the neural network. The activations and/or weights may be compressed based on a compression ratio or a system event. The system event may be a bandwidth condition, a power condition, a debug condition, a thermal condition or the like. The apparatus may operate the neural network to compute an inference based on the compressed activations or the compressed weights.

Abstract: A configurable neuro-inspired convolution processor is designed as an array of neurons each operating in an independent clock domain. The processor implements a recurrent network using efficient sparse convolutions with zero-patch skipping for feedforward operations, and sparse spike-driven reconstruction for feedback operations. A globally asynchronous locally synchronous structure enables scalable design and load balancing to achieve 22% reduction in power. Fabricated in 40 nm CMOS, the 2.56 mm2 inference processor integrates 48 neurons, a hub and an Open RISC processor. The chip achieves 718 GOPS at 380 MHz, and demonstrates applications in feature extraction from images and depth extraction from stereo images.

Abstract: A non-transitory computer-readable recording medium stores a learning program that causes a computer to execute a process including: extracting, from a plurality of data groups, a plurality of first data groups having an order; generating, for each data element corresponding to each of the first data groups, an ordered data matrix in which data elements having same order have value corresponding to relationship among the data elements and data elements having different orders have values corresponding to the different orders; and obtaining input tensor data by performing tensor decomposition with the ordered data matrix, inputting the input tensor data to a neural network at time of performing deep machine learning, performing deep machine learning of the neural network, and learning about method for the tensor decomposition.

Abstract: A neuromorphic chip includes synaptic cells including respective resistive devices, axon lines, dendrite lines and switches. The synaptic cells are connected to the axon lines and dendrite lines to form a crossbar array. The axon lines are configured to receive input data and to supply the input data to the synaptic cells. The dendrite lines are configured to receive output data and to supply the output data via one or more respective output lines. A given one of the switches is configured to connect an input terminal to one or more input lines and to changeably connect its one or more output terminals to a given one or more axon lines.

Abstract: A platform that integrates and collates the data points from students, employers, schools, and industry into an ecosystem which allows for customers (students, employers, schools, and industry) to model ‘what-if’ scenarios based on their industry parameters. By using a design algorithm based on automated reasoning, game theory, and knowledge mining, within a neural network, the platform can predict, model, and build the journey. The decision modeling neural learning platform may be used to augment or replace the need for guidance counselors in schools, along with assisting industry and immigration liaisons.

Abstract: Embodiments of the invention relate to a time-division multiplexed neurosynaptic module with implicit memory addressing for implementing a neural network. One embodiment comprises maintaining neuron attributes for multiple neurons and maintaining incoming firing events for different time steps. For each time step, incoming firing events for said time step are integrated in a time-division multiplexing manner. Incoming firing events are integrated based on the neuron attributes maintained. For each time step, the neuron attributes maintained are updated in parallel based on the integrated incoming firing events for said time step.

Abstract: A method of generating an outcome for a sporting event is disclosed herein. A computing system retrieves tracking data from a data store. The computing system generates a predictive model using a deep neural network. The one or more neural networks of the deep neural network generates one or more embeddings comprising team-specific information and agent-specific information based on the tracking data. The computing system selects, from the tracking data, one or more features related to a current context of the sporting event. The computing system learns, by the deep neural network, one or more likely outcomes of one or more sporting events. The computing system receives a pre-match lineup for the sporting event. The computing system generates, via the predictive model, a likely outcome of the sporting event based on historical information of each agent for the home team, each agent for the away team, and team-specific features.

Abstract: A time-series-data feature extraction device includes: a data processing unit that processes a received unevenly spaced time-series-data group into an evenly spaced time-series-data group including omissions and an omission information group indicating presence or absence of omissions, based on a received input time-series data length and a received minimum observation interval; a model learning unit that learns a weight vector of each layer of a model with a difference between an element not missing in a matrix of the evenly spaced time-series-data group including omissions and an element of an output result of an output layer of the model being taken as an error, and stores the weight vector as a model parameter in a storage unit, the difference being; and a feature extraction unit that receives time-series data of a feature extraction target, calculates a value of the intermediate layer of the model with use of the model parameter stored in the storage unit by inputting the received time-series data of the

Abstract: A synapse memory and a method for reading a weight value stored in a synapse memory are provided. The synapse memory includes a memory device configured to store a weight value. The memory device includes a read terminal, a write terminal, and a common terminal, the read terminal being configured to receive a read signal, the write terminal being configured to receive a write signal, and the common terminal being configured to output an output signal from the memory device. The synapse memory also includes a write transistor provided between the write terminal of the memory device and a write signal line configured to send the write signal. The synapse memory further includes a common transistor provided between the common terminal of the memory device and one of the dendrite lines.

Abstract: A histogram-based method of selecting a fixed point number format for representing a set of values input to, or output from, a layer of a Deep Neural Network (DNN). The method comprises obtaining a histogram that represents an expected distribution of the set of values of the layer, each bin of the histogram is associated with a frequency value and a representative value in a floating point number format; quantising the representative values according to each of a plurality of potential fixed point number formats; estimating, for each of the plurality of potential fixed point number formats, the total quantisation error based on the frequency values of the histogram and a distance value for each bin that is based on the quantisation of the representative value for that bin; and selecting the fixed point number format associated with the smallest estimated total quantisation error as the optimum fixed point number format for representing the set of values of the layer.

Abstract: The disclosure relates to an artificial intelligence (AI) system for performing functions, such as recognition, determination, and the like, of a human being using a neural network model such as deep learning, and an application thereof. A method, performed by a first electronic device, of processing a neural network model is provided. The method includes: acquiring an input value of the neural network model; processing at least one node included in a first group in the neural network model based on the input value; acquiring at least one value output from the first group by processing the at least one node included in the first group; identifying a second electronic device by which at least one node included in a second group in the neural network model is to be processed; and transmitting the at least one value output from the first group to the identified second electronic device.

Abstract: A synapse memory system includes: synapse memory cells provided at cross points of axon lines and dendrite lines, each synapse memory cell being associated with nonvolatile random-access memory (NVRAM), each synapse memory cell being configured to store a weight value according to an output level of a write signal; a write portion configured to write the weight value to each synapse memory cell, the write portion including a write driver and an output controller, the write driver being a digital driver configured to output the write signal to a subject synapse memory cell, the output controller being configured to control the output level of the write signal of the write driver; and read drivers configured to read the weight value stored in the synapse memory cells.

Abstract: Embodiments for identifying significant events for finding a root cause of an anomaly collecting time series data for events for each network device by detecting an anomaly in the time series data comprising an outlier on an edge of the time series data by comparing a predicted value of the event to an actual value of the event using a selected forecasting model; declaring the event to be an anomaly at a particular time if a difference between the predicted value and actual value exceed a defined threshold based on residual values for other devices; analyzing in a combined RNN/LSTM process all events for all devices of the network within a time proximity of the particular time of the anomaly to filter usual events and rank each event relative to the anomaly; and displaying a labeled chart of the time series data showing the anomaly in a graph relative to all the events.

Abstract: Techniques are provided which may allow an artificial intelligence modeling engine to be used for multiple applications. A user may configure models without detailed knowledge, which may allow broader use of artificial intelligence engines. Operators optimized in one model may be applied to data with similar attributes in another model.

Abstract: A method, computer program product, and apparatus for adapting a trained neural network having one or more batch normalization layers are provided. The method includes adapting only the one or more batch normalization layers using adaptation data. The method also includes adapting the whole of the neural network having the one or more adapted batch normalization layers, using the adaptation data.

Abstract: Systems and methods for private deep neural network training are disclosed. Method includes storing first private values at first machine and second private values at second machine; providing, to third machine, first share of first private values and first share of second private values; providing, to fourth machine, second share of first private values and second share of second private values; computing, at third machine, third machine-value based on first share of first private values and first share of second private values; computing, at fourth machine, fourth machine-value based on second share of first private values and second share of second private values; providing, to first machine and second machine, third machine-value and fourth machine-value; and computing, at first machine, a mathematical function of first private values and second private values, mathematical function being computed based on first private values stored at first machine, third machine-value, and fourth machine-value.

Abstract: Described is a system for pattern recognition designed for neuromorphic hardware. The system generates a spike train of neuron spikes for training patterns with each excitatory neuron in an excitatory layer, where each training pattern belongs to a pattern class. A spiking rate distribution of excitatory neurons is generated for each pattern class. Each spiking rate distribution of excitatory neurons is normalized, and a class template is generated for each pattern class from the normalized spiking rate distributions. An unlabeled input pattern is classified using the class templates. A mechanical component of an autonomous device can be controlled based on classification of the unlabeled input pattern.

Abstract: A circuit for performing neural network computations for a neural network comprising a plurality of layers, the circuit comprising: activation circuitry configured to receive a vector of accumulated values and configured to apply a function to each accumulated value to generate a vector of activation values; and normalization circuitry coupled to the activation circuitry and configured to generate a respective normalized value from each activation value.

Abstract: A non-transitory computer-readable recording medium stores therein a learning program that causes a computer to execute a process including: generating, from graph data subject to learning, extended graph data that has a value of each node included in the graph data, and a value corresponding to a distance between each node and another node included in the graph data; and obtaining input tensor data by performing tensor decomposition of the generated extended graph data, performing deep learning with a neural network by inputting the input tensor data into the neural network upon deep learning, and learning a method of the tensor decomposition.

Abstract: The present disclosure discloses a method and apparatus for scheduling a resource for a deep learning framework. The method can comprise: querying statuses of all deep learning job objects from a Kubernetes platform at a predetermined interval; and submitting, in response to finding from the queried deep learning job objects a deep learning job object having a status conforming to a resource request submission status, a resource request to the Kubernetes platform to schedule a physical machine where the Kubernetes platform is located to initiate a deep learning training task. The method can completely automate the allocation and release on the resource of the deep learning training task.

Abstract: A non-transitory computer-readable recording medium stores a learning program that causes a computer to execute a machine learning process for graph data. The machine learning process includes: generating, from graph data to be subjected to learning, extended graph data where at least some of nodes included in the graph data have a value of the nodes and a value corresponding to presence or absence of an indefinite element at the nodes; and obtaining input tensor data by performing tensor decomposition of the generated extended graph data, performing deep learning with a neural network by inputting the input tensor data into the neural network upon deep learning, and learning a method of the tensor decomposition.

Abstract: A training system (10) for a street vehicle (1) for detecting non-automobile road users (2) based on sounds, comprising an input interface (11) for receiving target training data (12), wherein the target training data (12) are audio recordings (17) of the non-automobile road user (2) recorded by at least one microphone (3a, 3b, 3c, 3d) located on the street vehicle (1) while driving the street vehicle (1), and respective associated target characteristics (18a) of this non-automobile road user (2), and wherein the training system (10) is configured to forward propagate an artificial neural network (13) with the target training data (12) and to record an actual characteristic (18b) of the respective non-automobile road user (2) determined with the artificial neural network (13) in the forward propagation, and to obtain weighting factors (14) for the connections (15) of neurons (16) in the artificial neural network (13) through backward propagation of the artificial neural network (13) with the difference (19) b

Abstract: A method of identifying a defensive alignment and an offensive alignment in a set-piece is disclosed herein. A computing system receives one or more streams of tracking data. The computing system identifies a set-piece contained in the one or more streams of tracking data. The computing system identifies a defensive alignment of a first team and an offensive alignment of a second team. The computing system extracts, via a convolutional neural network, one or more features corresponding to a type of defensive alignment implemented by the first team by passing the set-piece through the convolutional neural network. The computing system scans the set-piece, via a machine learning algorithm, to identify one or more features indicative of a type of offensive alignment implemented by the second team. The computing system infers the type of defensive alignment implemented by the first team.

Abstract: A processor-implemented data processing method includes encoding a plurality of weights of a filter of a neural network using an inverted two's complement fixed-point format; generating weight data based on values of the encoded weights corresponding to same filter positions of a plurality of filters; and performing an operation on the weight data and input activation data using a bit-serial scheme to control when to perform an activation function with respect to the weight data and input activation data.

Abstract: The present disclosure relates to a machine learning accelerator system and methods of transporting data using the machine learning accelerator system. The machine learning accelerator system may include a switch network comprising an array of switch nodes, and an array of processing elements. Each processing element of the array of processing elements is connected to a switch node of the array of switch nodes and is configured to generate data that is transportable via the switch node. The method may include receiving input data using a switch node from a data source and generating output data based on the input data, using a processing element that is connected to the switch node. The method may include transporting the generated output data to a destination processing element using a switch node.

Abstract: The present technology addresses the problem of quickly and safely improving policies in online reinforcement learning domains. As its solution, an exploration strategy comprising diverse exploration (DE) is employed, which learns and deploys a diverse set of safe policies to explore the environment. DE theory explains why diversity in behavior policies enables effective exploration without sacrificing exploitation. An empirical study shows that an online policy improvement algorithm framework implementing the DE strategy can achieve both fast policy improvement and safe online performance.

Abstract: A method, computer program product and system for generating a neural network. Initial neural networks are prepared, each of which includes an input layer containing one or more input nodes, a middle layer containing one or more middle nodes, and an output layer containing one or more output nodes. A new neural network is generated that includes a new middle layer containing one or more middle nodes based on the middle nodes of the middle layers of the initial neural networks.

Abstract: An apparatus of operating a computational network is configured to determine a low-rank approximation for one or more layers of the computational network based at least in part on a set of residual targets. A set of candidate rank vectors corresponding to the set of residual targets may be determined. Each of the candidate rank vectors may be evaluated using an objective function. A candidate rank vector may be selected and used to determine the low rank approximation. The computational network may be compressed based on the low-rank approximation. In turn the computational network may be operated using the one or more compressed layers.

Abstract: During training mode, first input data is provided to a first neural network to generate first output data indicating that the first input data is classified in a first cluster. The first input data includes at least one of a continuous feature or a categorical feature. Second input data is generated and provided to at least one second neural network to generate second output data. The at least one second neural network corresponds to a variational autoencoder. An aggregate loss corresponding to the second output data is determined, including at least one of evaluating a first loss function for the continuous feature or evaluating a second loss function for the categorical feature. Based on the aggregate loss, at least one parameter of at least one neural network is adjusted. During use mode, the neural networks are used to determine cluster identifications and anomaly likelihoods for received data samples.

Abstract: Systems and methods for unsupervised representation learning by sorting sequences are provided. An unsupervised representation learning approach is provided which uses videos without semantic labels. The temporal coherence as a supervisory signal can be leveraged by formulating representation learning as a sequence sorting task. A plurality of temporally shuffled frames (i.e., in non-chronological order) can be used as inputs and a convolutional neural network can be trained to sort the shuffled sequences and to facilitate machine learning of features by the convolutional neural network. Features are extracted from all frame pairs and aggregated to predict the correct sequence order. As sorting shuffled image sequence requires an understanding of the statistical temporal structure of images, training with such a proxy task can allow a computer to learn rich and generalizable visual representations from digital images.

Abstract: An information processing apparatus includes a memory that stores information indicating a first individual and a processing result that is output by each node of the first individual in a process of executing image processing based on the first individual; and a processor coupled to the memory and configured to: generate a second individual based on the first individual; specify subtrees that have the same content and include terminating nodes between a tree structure indicating the second individual and a tree structure indicating the first individual; and set a processing result corresponding to a head node of the subtree included in the first individual, which is stored in the memory, as a result of executing image processing based on the subtree included in the second individual when adaptability of the second individual is calculated based on a result of executing image processing based on the second individual.

Abstract: A system includes a processor that executes instructions stored in a memory to implement an auto-solution advisor on a server. The auto-solution advisor receives a current help request describing in lay language text a problem with a computer application or computing device, and determines whether the current help request is textually similar to a previous help request for a previous problem. Based on the similarity of the current help request to the previous help request, the auto-solution advisor assigns a known solution for the previous problem as the suggested solution for the current help request.

Abstract: A system and method for predicting multi-agent locations is disclosed herein. A computing system retrieves tracking data from a data store. The computing system generates a predictive model using a conditional variational autoencoder. The conditional variational autoencoder learns one or more paths a subset of agents of the plurality of agents are likely to take. The computing system receives tracking data from a tracking system positioned remotely in a venue hosting a candidate sporting event. The computing system identifies one or more candidate agents for which to predict locations. The computing system infers, via the predictive model, one or more locations of the one or more candidate agents. The computing system generates a graphical representation of the one or more locations of the one or more candidate agents.

Abstract: A cognitive information processing system environment which includes a plurality of data sources; a cognitive inference and learning system coupled to receive a data from the plurality of data sources, the cognitive inference and learning system processing the data from the plurality of data sources to provide cognitively processed insights, the cognitive inference and learning system further comprising performing a learning operation to iteratively improve the cognitively processed insights over time; and, a destination, the destination receiving the cognitively processed insights.

Abstract: An approach is provided for a redundant feature detection engine. The approach, for instance, involves segmenting an input image into a plurality of grid cells for processing by the redundant feature detection engine. The redundant feature detection engine includes a neural network. The approach also involves, for each of the plurality of grid cells, initiating a prediction of an object code by the redundant feature detection engine. The object code is a predicted feature that uniquely identifies an object depicted in the input image. The approach further involves aggregating the plurality of grid cells into one or more clusters based on the object code predicted for said each grid cell. The approach further involves predicting one or more features of the object corresponding to a respective cluster of the one or more clusters by merging one or more feature prediction outputs of said each grid cell in the respective cluster.