Abstract: A computer-implemented method, according to one implementation, includes: monitoring requests received for an AI interface prompt in real-time, and determining whether one or more of the requests violate compliance metrics. Risk scores are calculated for requests determined as violating the compliance metrics. The requests determined as violating the compliance metrics are updated by implementing protective measures correlated with the calculated risk scores. Moreover, the updated requests are sent to the AI interface prompt.

Abstract: Embodiments are directed to the tracking of data in a generative adversarial network (GAN) model using a distributed ledger system, such as a blockchain. A learning platform implementing a classification model receives, from a third party, a set of data examples generated by a generator model. The set of data examples are processed by the classification model, which outputs a prediction for each data example indicating whether each data example is true or false. The distributed ledger keeps a record of data examples submitted to the learning platform, as well as of predictions determined by the classification model on the learning platform. The learning platform analyzes the records of the distributed ledger, and pairs the records corresponding to the submitted data examples and the generated predictions determined by the classification model, and determines if the predictions were correct. The classification model may then be updated based upon the prediction results.

Abstract: Systems and methods of the present disclosure enable a processor to automatically detect duplicate data entries by receiving data entries associated with a user, where each data entry includes a value, a time, an entity identifier, and a location. Pairs of similar data entries are determined by matching the entity identifier and the location pairs data entries. Candidate duplicate data entries are determined based on a proximity in time between data entries of the similar data entries. For each candidate duplicate data entry, a feature vector is generated including the entity identifier, location, value and time, and each feature vector is submitted to a duplicate classification model to automatically determine duplicate data entries from the candidate duplicate data entries, the duplicate classification model being trained according to a historical dispute entries.

Abstract: An interactive interpretation session with respect to a first version of a machine learning model is initiated. In the session, indications of factors contributing to a prediction decision are provided, as well indications of candidate model enhancement actions. In response to received input, an enhancement action is implemented to obtain a second version of the model. The second version of the model is stored.

Abstract: Concepts and technologies disclosed herein are directed to machine learning model understanding as-a-service. According to one aspect of the concepts and technologies disclosed herein, a model understanding as-a-service system can receive, from a user system, a service request that includes a machine learning model created for a user associated with the user system. The model understanding as-a-service system can conduct an analysis of the machine learning model in accordance with the service request. The model understanding as-a-service system can compile, for the user, results of the analysis of the machine learning model in accordance with the service request. The model understanding as-a-service system can create a service response that includes the results of the analysis. The model understanding as-a-service system can provide the service response to the user system.

Abstract: Some embodiments provide a method for training a machine-trained (MT) network that processes input data using network parameters. The method maps a set of input instances to a set of output values by propagating the set of input instances through the MT network. The set of input instances include input instances for each of multiple categories. The method selects multiple input instances as anchor instances. For each anchor instance, the method computes a loss function as a comparison between the output value for the anchor instance and each output value for an input instance in a different category than the anchor. The method computes a total loss function for the MT network as a sum of the loss function computed for each anchor instance. The method trains the network parameters using the computed total loss function.

Abstract: A computer-implemented method, a computer program product, and a computer processing system are provided for online learning for a Dynamic Boltzmann Machine (DyBM) with hidden units. The method includes imposing, by a processor device, limited connections in the DyBM where (i) a current observation x[t] depends only on latest hidden units h[t-1/2] and all previous observations x[<t] and (ii) the latest hidden units h[t-1/2] depend on all the previous observations x[<t] while being independent of older hidden units h[t-1/2]. The method further includes computing, by the processor device, gradients of an objective function. The method also includes optimizing, by the processor device, the objective function in polynomial time using a stochastic Gradient Descent algorithm applied to the gradients.

Abstract: An information processing method implemented by a computer includes: obtaining a piece of first data, and a piece of second data not included in a training dataset for training an inferencer; calculating, using a piece of first relevant data obtained by inputting the first data to the inferencer trained by machine learning using the training dataset, a first contribution representing contributions of portions constituting the first data to a piece of first output data output by inputting the first data to the inferencer; calculating, using a piece of second relevant data obtained by inputting the second data to the inferencer, a second contribution representing contributions of portions constituting the second data to a piece of second output data output by inputting the second data to the inferencer; and determining whether to add the second data to the training dataset, according to the similarity between the first and second contributions.

Abstract: Computer-implemented techniques encompass using distinct machine learning sub-models to score respective types of candidate content for the purpose of providing personalized content suggestions to end-users of a content management system. The relevancy scores generated by the distinct sub-models are mapped to expected end-user interaction scores of the candidate content scored. Content suggestions are provided at end-users' computing devices where the suggested content is selected from the candidate content based on the expected end-user interaction scores of the candidate content. For each distinct sub-model, a normalizing mapping function is solved using an optimizer that maps the relevancy scores generated by the sub-model for the candidate content to expected end-user interaction scores for the candidate content. The expected end-user interaction scores are comparable across the distinct sub-models and can be used to rank content suggestions across the distinct sub-models.

Abstract: A system uses neural networks to determine intents of traffic entities (e.g., pedestrians, bicycles, vehicles) in an environment surrounding a vehicle (e.g., an autonomous vehicle) and generates commands to control the vehicle based on the determined intents. The system receives images of the environment captured by sensors on the vehicle, and processes the images using neural network models to determine overall intents or predicted actions of the one or more traffic entities within the images. The system generates commands to control the vehicle based on the determined overall intents of the traffic entities.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for training a neural network used to generate a ranking score for a network input. One of the methods includes generating training data and training the neural network on the training data. The training data includes a plurality of training pairs. The generating comprising: obtaining data indicating that a plurality of training network inputs were displayed in a user interface according to a presentation order, obtaining data indicating that a first training network input of the plurality of training network inputs has a positive label, determining that a second training network input of the plurality of training network inputs (i) has a negative label and (ii) is higher than the first training network input in the presentation order, and generating a training pair that includes the first training network input and the second training network input.

Abstract: Methods, systems, and computer-readable media are disclosed herein for a concurrent comparative tool for assessing sub-models of a data model pipeline in a deployed or pre-deployment environment. The tool may compute a plurality of performance measures that quantitatively assess the performance of each sub-model in the data model pipeline based on a configuration file that facilitates validation of the technological performance and predictive accuracy of the sub-model. Additionally, multiple versions of a sub-model deployed in similar data model pipelines, or in a pre-deployment environment, may be comparatively evaluated. A leading version of the sub-model may be identified and deployed.

Abstract: A method for detecting an overlay precision and a method for compensating an overlay deviation are provided.

Abstract: The present teaching relates to method, system, medium, and implementation of in-situ perception in an autonomous driving vehicle. A plurality of types of sensor data are acquired continuously via a plurality of types of sensors deployed on the vehicle, where the plurality of types of sensor data provide information about surrounding of the vehicle. One or more items surrounding the vehicle are tracked, based on some models, from a first of the plurality of types of sensor data from a first type of the plurality of types of sensors. A second of the plurality of types of sensor data are obtained from a second type of the plurality of sensors and are used to generate validation base data. Some of the one or more items are labeled, automatically, via validation base data to generate labeled at least some item, which is to be used to generate model updated information for updating the at least one model.

Abstract: Systems and methods for managing change requests are disclosed. A system for managing change requests may include a memory storing instructions and at least one processor configured to execute instructions to perform operations including: receiving, from a client device, a change request; routing the change request to a first similarity determination pipeline, based on the first classification, identifying an implementation device; and transmitting the change request to the implementation device. The first similarity determination pipeline may be configured to: extract at least one first request element from the change request; determine a first group of change requests based on the at least one first extracted request element; determine a first similarity metric between the change request and the first group of change requests; and determine a first classification of the change request based on the first similarity metric.

Abstract: An online concierge shopping system identifies candidate items to a user for inclusion in an order based on prior user inclusion of items in orders and items currently included in the order. From a multi-dimensional tensor generated from cooccurrences of items in orders from various users, the online concierge system generates item embeddings and user embeddings in a common latent space by decomposing the multi-dimensional tensor. From items included in an order, the online concierge system generates an order embedding from item embeddings of the items included in the order. Scores for candidate items are determined based on similarity of item embeddings for the candidate items to the order embedding. Candidate items are selected based on their scores, with the selected candidate items identified to the user.

Abstract: Provided herein are systems and methods for an iterative approach to topic modeling and the use of web mapping technology to implement advanced spatial operators for interactive high-dimensional visualization and inference.

Abstract: A learning system includes a learning data acquisition unit that generates a neural network included in a single forecast model for performing a forecast of demand for a service through machine learning and acquires, as learning data, an actual demand value of the service and a feature amount for learning for each mesh associated with a past period, the feature amount for learning including the number of demands for the service in each mesh in each period as a feature amount, a learning data normalization unit that normalizes at least the number of demands in the feature amount for learning for each mesh, and a generation unit that performs machine learning on the basis of an actual demand value and a demand forecast value obtained by inputting the feature amount for learning including the normalized number of demands to the neural network, and generates the neural network.

Abstract: Systems and methods for training a machine learning (“ML”) model for use in controlling an autonomous vehicle (“AV”) are described herein. Implementations can obtain an initial state instance from driving of a vehicle, obtain ground truth label(s) for subsequent state instance(s) that each indicate a corresponding action of the vehicle for a corresponding time instance, perform, for a given time interval, a simulated episode, of locomotion of a simulated AV, generate, for each of a plurality of time instances of the given time interval, subsequent simulated state instance(s) that differ from the subsequent state instance(s), determine, using the ML model, and for each of the time instances, a predicted simulated action of the simulated AV based on the subsequent simulated operation instance(s), generate loss(es) based on the predicted simulated actions and the ground truth labels, and update the ML model based on the loss(es).

Abstract: An example method embodying the disclosed technology comprises: digitally storing Teacher models and a Student model at a server computer; training each model with a corpus of unlabeled training data using Masked Language Modeling; fine-tuning each Teacher model for an ASAG task with labeled ground truth data; executing each Teacher model to generate and digitally store a respective set of class probabilities on an unlabeled task-specific data set for the ASAG task; further training the Student model by a linear ensemble of the Teacher models using KD; receiving, at the server computer, digital input comprising a target response text and a corresponding target reference answer text; programmatically inputting the target response text and the corresponding target reference answer text to the Student model, thereby outputting a corresponding predicted binary label; displaying correction data indicating the corresponding predicted binary label in a GUI; and, optionally, displaying explainability data in the GUI.

Abstract: A device for self-adjusting an electrical threshold includes a framing unit that compares a setpoint signal representative of the supply signal to a plurality of successive intervals of reference signals that increase in value according to a geometric sequence. The framing unit generates an output based on the interval bounding the setpoint signal, wherein the output controls a switch that is configured to provide one of the plurality of reference signals as an output, wherein the output reference signal is representative of the electrical detection threshold.

Abstract: A computer-implemented system, platform, method and computer program product for optimizing a data analytics fraud prediction/detection pipeline that includes a combination of a classical machine learned classifier model with a quantum machine learned model to optimize the performance of the fraud prevention model. The feature selection uses different feature maps: one determined by the classic classifier and the other determined by the quantum model implementation that exploits the entanglement quantum property. The quantum method can include a quantum support vector machine implementing a built feature forward algorithm that uses a quantum kernel estimate for feature mapping. This quantum model can be run on a quantum computer or quantum simulator that can run a quantum algorithm built for extracting feature importance.

Abstract: Described are systems and methods for generating recommendation campaigns that optimize for both a desired short-term user behavior and a desired long-term user behavior. In comparison to existing techniques that focus on targeting advertisements or recommendations to specific individuals with a single goal of receiving an interaction with the advertisement from that individual (i.e., a desired short-term behavior), the disclosed implementations consider the long-term user behavior, such as increased visits to a website during a long-term rage, and generate a recommendation campaign that also optimizes for that desired long-term user behavior.

Abstract: The present disclosure relates to identifying process flows from log sources (e.g., log files), and generating visual representations (e.g., flow diagrams, Sankey diagrams, etc.) of the identified process flows. In addition, the present disclosure relates to clustering of tree structures based on the shape of the tree structure using one or more hashing algorithms. The present disclosure also relates to a user interface that presents a query builder for efficiently querying a log analytics system for tree structures that satisfy a user-defined range.

Abstract: A computing platform is configured to: (i) at a first time, input data values for a first set of data variables associated with a given construction project into a first machine-learning model that functions to output a prediction of a first set of reference projects that are similar to the given construction project, (ii) based on historical data for the first set of reference projects, determine a predicted value for a parameter of the given construction project, (iii) at a second time, input data values for a second set of data variables associated with the given construction project into a second machine-learning model that functions to output a prediction of a second set of reference projects that are similar to the given construction project, and (iv) based on historical data for the second set of reference projects, determine an updated predicted value for the parameter of the given construction project.

Abstract: In one or more embodiments, one or more methods, processes, and/or systems may receive data associated with completion of tasks by agents. Each task corresponds to a category of tasks and is associated with an outcome relative to satisfaction of a specification of performance by a work distributor. An aptitude prediction model is trained to map, for each category of task, a correlation between an outcome corresponding to satisfaction of the specification of performance and one or more aspects of each task and one or more attributes of each agent that has completed the task. An aptitude of each agent towards a category of tasks is determined. A probability that a first agent will complete a first task in a manner specified by the work distributor is predicted using the trained aptitude prediction model. Identification information of the first agent is provided for display in association with the determined probability.

Abstract: The disclosure relates to a video anomaly detection method based on human-machine cooperation, in which video frames and traditional descriptors of optical stream of an image are utilized as an input for auto-encoder neural network coding, and converted into a representation content of a hidden layer, and then the representation content of the hidden layer is decoded, reconstructed and output. The auto-encoder network is trained with normal samples. In a test stage, if an input is a normal sample, a final reconstructed error keeps high similarity with an input sample; on the contrary, if the input is an abnormal sample, the final reconstructed error deviates greatly from the input sample.

Abstract: A message suggestion service may use clusters of pre-approved messages to improve the quality of messages suggested to users. During a conversation, messages of the conversation may be processed with a neural network to compute a conversation encoding vector. The neural network may also be used to compute pre-approved message encoding vectors of the pre-approved messages. Distances between the conversation encoding vector and the pre-approved message encoding vectors may be used to select one or more clusters. Distances between the conversation encoding vector and the pre-approved message encoding vectors may then be used to select one or more pre-approved messages from the selected clusters. The selected pre-approved messages may then be presented as suggested messages to a user.

Abstract: A method including: receiving, as input, an image; providing a neural network structure including a plurality of multilayer multi-scale neural networks, wherein the plurality of multilayer multi-scale neural networks are arranged sequentially, by laterally connecting corresponding scale-level layers between each two adjoining multilayer multi-scale neural networks in the sequence; and at a training stage, training the neural network structure on a training dataset, to obtain a trained machine learning model configured to perform a computer vision task which includes outputting at least one of: (i) a classification of the image into one class of a set of two or more classes, (ii) a segmentation of a least one object in the image, and (iii) a detection of at least one object in the image.

Abstract: A method and system for aggregating into a unique aggregated group (AG), protection groups (PGs) that are possible classifications with at least a threshold probability for a same unique combination of IP addresses. The PGs and the unique combination of IP addresses are included in the AG. Each of the IP addresses of the unique combination of IP addresses have respective associated probabilities for each PG included in the AG. The method further includes selecting and providing for display AGs based on the probabilities associated with the respective IP addresses included in the AGs, and providing for display at least one interactive graphical element in association with each AG selected for display. User activation of one of the interactive graphical element accepts assignment of one or more selected IP addresses included in the AG to a selected one of the one or more PGs included in the AG.

Abstract: A method includes receiving, with a computing device, a first client request from a first client that identifies a machine learning model and a sensor. The method includes sending, with the computing device, a call to a server to apply the identified machine learning model to a set of data from the identified sensor, in response to the first client request. The method includes receiving, with the computing device, a second client request from a second client that identifies a same machine learning model and sensor as the first client request. The method includes sending, with the computing device, response data from the identified machine learning model to both the first client and the second client without sending an additional call to the server in response to the second client request.

Abstract: Certain aspects of the present disclosure provide techniques for performing finite rank deep kernel learning. In one example, a method for performing finite rank deep kernel learning includes receiving a training dataset; forming a set of embeddings by subjecting the training dataset to a deep neural network; forming, from the set of embeddings, a plurality of dot kernels; linearly combining the plurality of dot kernels to form a composite kernel for a Gaussian process; receiving live data from an application; and predicting a plurality of values and a plurality of uncertainties associated with the plurality of values simultaneously using the composite kernel.

Abstract: A quantum computing service may store, in a cache, one or more compiled files of respective quantum functions included in one or more quantum computing programs received one or more customers. When the quantum computing service receives another quantum computing program, from the same or a different customer, the quantum computing service may determine whether the quantum computing program may include one or more of the quantum functions corresponding to the compiled files in the cache. If so, the quantum computing service may use the compiled files in the cache to compile the quantum computing program.

Abstract: An embodiment for dynamic categorization of information technology (IT) service tickets is provided. The embodiment may include logging an IT service ticket when text is entered into a description field. The embodiment may also include creating a filtered description field by processing the text entered into the description field. The embodiment may further include computing a set of exponential weights and assigning the set of exponential weights to each word in the filtered description field. The embodiment may also include multiplying the set of exponential weights by the word's TF-IDF score to determine an IT service ticket category for placement of the IT service ticket into the IT service ticket category. The embodiment may further include generating features for machine learning, utilizing the generated features to build a supervised machine learning model, and evaluating the supervised machine learning model through analyzation of data from historical IT service tickets.

Abstract: Techniques of data flow control are disclosed herein. One example technique includes upon receiving a notification indicating a change to a content item in a source shard, parsing the content item to extract data representing attributes of the content item and identifying a partition of the system-wide index based on the extracted data representing the attributes of the content item. The example technique can also include transmitting, to a token issuer, a request for a token that represents a permission to write to the identified partition of the system-wide index and upon receiving a token issued by the token issuer in response to the transmitted request, transmitting the extracted data representing the attributes of the content item along with the received token to write the extracted data in the partition of the system-wide index.

Abstract: Systems and methods for tracking velocity information. One system includes an application execution server providing an application layer. The application execution server is configured to receive a request including metadata. The application execution server is also configured to generate and transmit a response to the request. The application execution server is also configured to enrich the metadata by structuring the metadata for further processing by a data processing layer, where the further processing includes determining velocity information associated with the metadata, and by supplementing the metadata with available historical velocity information. The application execution server is also configured to transmit the enriched metadata for further processing by the data processing layer.

Abstract: A system for managing a client request is described herein, which may have at least one processor and a non-transitory computer-readable medium containing a set of instructions executable by the at least one processor. Execution of these instructions may cause the processor to perform steps of: validating a client request received from a remote client device, the client request including request data; transmitting, based on the validating, a response to the remote client device; based on the request data, determining a queue for the client request; asynchronously enqueuing the client request in the queue, the queue being configured to analyze the client request according to a model; analyzing the client request; and based on analyzing the client request, performing a responsive action.

Abstract: A copy of a model comprising a plurality of trees is received, as is a copy of training set data comprising a plurality of training set examples. For each tree included in the plurality of trees, the training set data is used to determine which training set examples are classified as a given leaf. A blame forest is generated at least in part by mapping each training set item to the respective leaves at which it arrives.

Abstract: In some examples, a system for decorating network traffic flows with outlier scores includes a processor and a memory device to store traffic flows received from a network. The processor is configured to receive a set of traffic flows from the memory device and generate a tree model to split the traffic flows into clusters of traffic flows. Each cluster corresponds with a leaf of the tree model. The processor is further configured to generate machine learning models for each of the clusters of traffic flows separately. For a new traffic flow, the processor is configured to identify a specific one of the machine learning models that corresponds with the new traffic flow, compute an outlier score for the new traffic flow using the identified specific one of the machine learning models, and decorate the new traffic flow with the outlier score.

Abstract: A first set of features associated with a neural network are parameterized. A decision tree is generated from the first set of features. One or more adjustments for the neural network are received at the decision tree. A second set of features associated with the adjustments at the decision tree are parameterized. The parameterized first and second set of features are combined into a plurality of parameters. From the plurality, an adjusted neural network is generated.

Abstract: Implementations are described herein for analyzing existing interactive web sites to facilitate automatic engagement with those web sites, e.g., by automated assistants or via other user interfaces, with minimal effort from the hosts of those websites. For example, in various implementations, techniques described herein may be used to extract, validate, maintain, generalize, extend and/or distribute individual actions and “traces” of actions that are useable to navigate through various interactive websites. Additionally, techniques are described herein for leveraging these actions and/or traces to automate aspects of interaction with a third party website.

Abstract: A system and methods to identify which signals are significant to an assessment of a complex machine system state in the presence of non-linearities and disjoint groupings of condition types. The system enables sub-grouping of signals corresponding to system sub-components or regions. Explanations of signal significance are derived to assist in causal analysis and operational feedback to the system is prescribed and implemented for the given condition and causality.

Abstract: Aspects of the present disclosure describe distributed fiber optic sensing (DFOS) systems, methods, and structures that employ a distributed fiber optic sensor placement procedure that advantageously provides a desirable sensor coverage over a network at minimal cost.

Abstract: A system is provided for facilitating multi-label classification. During operation, the system maintains a set of training vectors. A respective vector represents an object and is associated with one or more labels that belong to a label set. After receiving an input vector, the system determines a similarity value between the input vector and one or more training vectors. The system further determines one or more labels associated with the input vector based on the similarity values between the input vector and the training vectors and their corresponding associated labels.

Abstract: Disclosed are systems and methods for predicting patient response to a treatment option. In one embodiment, the image slides from patient tissue samples are divided into patches and morphological patterns correlated with a disease outcome are labeled and given a patch-level score, based on whether the morphological patterns occur only in patients with good outcomes or patients with poor outcomes. A patient-level score can be generated based, at least partly, on the patch-level scores. Patch-level scores can identify regions of interest for targeted biomarker identification.

Abstract: This application describes systems and methods for generating machine learning models (MLMs). An exemplary method includes obtaining a sample and user input data characterizing a product or service. A subset of the data is selected from the sample based on sampling the sample according to the user input data. An MLM is trained by applying the data subset as training input to the MLM, thereby providing a trained MLM to emulate a customer selection process unique to the product or service. A user interface (UI) configured to receive other user input data and cause the trained MLM to execute on the other user input data, thereby testing the trained MLM, is presented. A summary of results from the execution of the trained MLM is generated and presented in the UI. The summary of results indicates a contribution to the trained MLM of each of a plurality of features.

Abstract: The present disclosure provides a fluorescence image registration method, the method includes: acquiring a fluorescence image of a biochip; selecting a preset local region of the fluorescence image; acquiring a position of a minimum value of a sum of brightness values of pixels in a first direction and a second direction, and obtaining pixel-level registration points; dividing the pixel-level registration points into non-defective pixels and defective pixels; if the fluorescence image meets the preset standard, correcting positions of the defective pixels according to positions of the non-defective pixels; acquiring a position of a center of gravity of image points of fluorescent molecules according to a center of gravity method; fitting straight lines in the first direction and the second direction respectively according to the position of the center of gravity; and acquiring boundary points of the fluorescence image and calculating the positions of the boundary points.

Abstract: Techniques are generally described for machine learning exampled-based annotation of image data. In some examples, a first machine learning model may receive a query image comprising a first depiction of an object-of-interest. In some examples, the first machine learning model may receive a target image representing a scene in which a second depiction of the object-of-interest is visually represented. In various examples, the first machine learning model may generate annotated output image data that identifies a location of the second depiction of the object-of-interest within the target image. In some examples, an object detection model may be trained based at least in part on the annotated output image data.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for training and deploying machine-learned communication. One of the methods includes: receiving an RF signal at a signal processing system for training a machine-learning network; providing the RF signal through the machine-learning network; producing an output from the machine-learning network; measuring a distance metric between the signal processing model output and a reference model output; determining modifications to the machine-learning network to reduce the distance metric between the output and the reference model output; and in response to reducing the distance metric to a value that is less than or equal to a threshold value, determining a score of the trained machine-learning network using one or more other RF signals and one or more other corresponding reference model outputs, the score indicating an a performance metric of the trained machine-learning network to perform the desired RF function.

Abstract: Systems and methods for automated, degradation-resistant, tuning of machine-learning (“ML”) models are provided. The systems and methods may identify an inoperative input utterance, retrieve a feature set associated with the inoperative input utterance, and generate an updated utterance-feature-intent (“UFI”) mapping based on the retrieved feature set. The systems and methods may retrain the ML model using the updated UFI mapping, and compare the accuracy of the system after the retraining and before the retraining. In a scenario where the accuracy of the system does not improve, the systems and methods may amplify the updated UFI mapping. In a scenario where the accuracy of the system does improve, the systems and methods may deploy the updated UFI mapping.