Abstract: A surgical instrument navigation system is provided that visually simulates a virtual volumetric scene of a body cavity of a patient from a point of view of a surgical instrument residing in the cavity of the patient. The surgical instrument navigation system includes: a surgical instrument; an imaging device which is operable to capture scan data representative of an internal region of interest within a given patient; a tracking subsystem that employs electro-magnetic sensing to capture in real-time position data indicative of the position of the surgical instrument; a data processor which is operable to render a volumetric, perspective image of the internal region of interest from a point of view of the surgical instrument; and a display which is operable to display the volumetric perspective image of the patient.

Abstract: An image processing method includes a first step of acquiring input data including a captured image and optical system information relating to a state of an optical system used for capturing the captured image and a second step of inputting the input data to a machine learning model and of generating an estimated image acquired by sharpening the captured image or by reshaping blurs included in the captured image.

Abstract: A method that may include training a student ODNN to mimic a teacher ODNN. The training may include calculating a teacher student detection loss that is based on a pre-bounding-box output of the teacher ODNN. The pre-bounding-box output of the teacher ODNN is a function of pre-bounding-box outputs of different ODNNs that belong to the teacher ODNN. The method may also include detecting one or more objects in an image, by feeding the image to the trained student ODNN; outputting by the trained student ODNN a student pre-bounding-box output; and calculating one or more bounding boxes based on the student pre-bounding-box output.

Abstract: Systems and methods for aggregating data. The system is configured to receive metadata from an interactive graphical user interface (GUI) of a user device, aggregate field values from the data stored on one or more databases based on the received metadata and generate filter instructions based on the received metadata. The system is further configured to transmit the aggregated field values and the filter instructions to the user device, receive a user-customized filter set and subscription request for a synthetic symbol associated with the user-customized filter set from the user device, and create the synthetic symbol responsive to the subscription request. Moreover, the system aggregates one or more data values from the data stored on the databases associated with the created synthetic symbol and generates instructions to display the data values on the interactive GUI in accordance with the user-customized filter set associated with the created synthetic symbol.

Abstract: Disclosed is a system for generating personalized recommendations based on dynamic and customized content selections and modeling of the content selections. The system may receive a request with an identifier and a query, and may obtain a particular recommendation configuration based the identifier and the query. The system may retrieve a set of content that satisfies the query and that is identified with at least one content prioritization parameter specified in the particular recommendation configuration, may generate a set of models of one or more model types that model relevance between the set of content and a different event specified in the particular recommendation configuration, and may compute a score for each content in each model based on the modeled relevance. The system may present recommended content in a different order than the set of content based on aggregate scores compiled for each content from the set of models.

Abstract: Techniques and apparatuses are described for machine-learning architectures for simultaneous connection to multiple carriers. In implementations, a network entity determines at least one deep neural network (DNN) configuration for processing information exchanged with a user equipment (UE) over a wireless communication system using carrier aggregation that includes at least a first component carrier and a second component carrier. At times, the at least one DNN configuration includes a first portion for forming a first DNN at the network entity, and a second portion for forming a second DNN at the UE. The network entity forms the first DNN based on the first portion and communicates an indication of the second portion to the UE. The network entity directs the UE to form the second DNN based on the second portion, and uses the first DNN to exchange, over the wireless communication system, the information with the UE using the carrier aggregation.

Abstract: An inference system trains and performs inference using a sparse neural network. The sparse neural network may include one or more layers, and each layer may be associated with a set of sparse weights that represent sparse connections between nodes of a layer and nodes of a previous layer. A layer output may be generated by applying the set of sparse weights associated with the layer to the layer output of a previous layer. Moreover, the one or more layers of the sparse neural network may generate sparse layer outputs. By using sparse representations of weights and layer outputs, robustness and stability of the neural network can be significantly improved, while maintaining competitive accuracy.

Abstract: A system and method for collecting and processing sensor data for facilitating and/or enabling autonomous, semi-autonomous, and remote operation of a vehicle, including: collecting surroundings at one or more sensors, and determining properties of the surroundings of the vehicle and/or the behavior of the vehicle based on the surroundings data at a computing system.

Abstract: A learning method executed by a computer, the learning method including augmenting original training data based on non-stored target information included in the original training data to generate a plurality of augmented training data, generating a plurality of intermediate feature values by inputting the plurality of augmented training data to a learning model, and learning a parameter of the learning model such that, with regard to the plurality of intermediate feature values, each of the plurality of intermediate feature values generated from a plurality of augmented training data, augmented from reference training data, becomes similar to a reference feature value.

Abstract: A data processing system digitally processes data feeds of inhaler device operation. The data feed represents operation of an inhaler device. The system indexes the live data feed with a key value representing the inhaler device for which the live data feed is obtained. For a particular key value indexed in the in-memory data storage, the system queries, a data feed representing physical operation of an inhaler device, segments the live data feed for that particular key value into a plurality of data samples, process at least a portion of the data samples to classify each of the processed data samples; outputs a prompt specifying whether operation of the inhaler device is within a threshold range of operation. Audio data, temperature data, image data, and ranging data can be processed to classify operation of the inhaler device and the order of operations of the inhaler device.

Abstract: A method for training a machine learning model, e.g., a neural network, using a regularization scheme is disclosed. The method includes generating regularized partial gradients of losses computed using an objective function for training the machine learning model.

Abstract: A process for training and sharing generic functional modules across multiple diverse (architecture, task) pairs for solving multiple diverse problems is described. The process is based on decomposing the general multi-task learning problem into several fine-grained and equally-sized subproblems, or pseudo-tasks. Training a set of (architecture, task) pairs then corresponds to solving a set of related pseudo-tasks, whose relationships can be exploited by shared functional modules. An efficient search algorithm is introduced for optimizing the mapping between pseudo-tasks and the modules that solve them, while simultaneously training the modules themselves.

Abstract: Tracking a current and/or previous position, velocity, acceleration, and/or heading of an object using sensor data may comprise determining whether to associate a current object detection generated from recently received (e.g., current) sensor data with a previous object detection generated from formerly received sensor data. In other words, a track may identify that an object detected in former sensor data is the same object detected in current sensor data. However, multiple types of sensor data may be used to detect objects and some objects may not be detected by different sensor types or may be detected differently, which may confound attempts to track an object. An ML model may be trained to receive outputs associated with different sensor types and/or a track associated with an object, and determine a data structure comprising a region of interest, object classification, and/or a pose associated with the object.

Abstract: Apparatus, device, methods and system relating to a vehicular telemetry environment for monitoring vehicle components and providing indications towards the condition of the vehicle components and providing optimal indications towards replacement or maintenance of vehicle components before vehicle component failure.

Abstract: A multi-layer technology stack includes a sensor layer including image sensors, a device layer, and a cloud layer, with interfaces between the layers. A method to curate different custom workflows for multiple applications include the following. Requirements for custom sets of data packages for the applications is received. The custom set of data packages include sensor data packages (e.g., SceneData) and contextual metadata packages that contextualize the sensor data packages (e.g., SceneMarks). Based on the received requirements and capabilities of components in the technology stack, the custom workflow for that application is deployed. This includes a selection, configuration and linking of components from the technology stack. The custom workflow is implemented in the components of the technology stack by transmitting workflow control packages directly and/or indirectly via the interfaces to the different layers.

Abstract: Methods, systems, and computer program products for analyzing one or more perceived or technical problems or proposed solutions, and proposing a result are disclosed. In accordance therewith, a query is received as an input, one or more documents that are most closely semantically related to the query are retrieved, a set of concept terms derived from each of the query and the retrieved semantically related documents is obtained, a list of generic Solution Prompts, each of which generic Solution Prompt thereof includes a placeholder for insertion of a word or phrase from the set of concept terms, is provided, and a morphological analysis is applied to combine the list of generic Solution Prompts with the obtained set of concept terms to create a list of Specific Solution Prompts.

Abstract: A workflow support apparatus includes a classification section that classifies a document included in an original document from image data acquired by reading the original document, and a workflow searching section that searches for a workflow to which the document is to be attached, from the document classified by the classification section.

Abstract: The present disclosure provides systems and methods that learn a loss function that, when (approximately) minimized over the training data, produces a model that performs well on test data according to some error metric. The error metric need not be differentiable and may be only loosely related to the loss function. In particular, the present disclosure presents a convex-programming-based algorithm that takes as input observed data from training a small number of models and produces as output a loss function. This algorithm can be used to tune loss function hyperparameters and/or to adjust the loss function on-the-fly during training.

Abstract: Methods, apparatus, systems and articles of manufacture are disclosed that optimize workflows. An example apparatus includes an intent determiner to determine an objective of a user input, the objective indicating a task to be executed in an infrastructure, a configuration composer to compose a plurality of workflows based on the determined objective, a model executor to execute a machine learning model to create a confidence score relating to the plurality of workflows, and a workflow selector to select at least one of the plurality of workflows for execution in the infrastructure, the selection of the at least one of the plurality of workflows based on the confidence score.

Abstract: Techniques are disclosed that enable automating user interface input by generating a sequence of actions to perform a task utilizing a multi-agent reinforcement learning framework. Various implementations process an intent associated with received user interface input using a holistic reinforcement policy network to select a software reinforcement learning policy network. The sequence of actions can be generated by processing the intent, as well as a sequence of software client state data, using the selected software reinforcement learning policy network. The sequence of actions are utilized to control the software client corresponding to the selected software reinforcement learning policy network.

Abstract: A search point determining method in an estimation process of a function, executed by a processor included in a search point determining apparatus, the method includes, calculating a search prediction time and a confidence interval upper limit obtained by using a Gaussian process for the function in each search candidate point from a past search result of the function, generating an area in a parameter space for each search candidate point by using a position of a search point close to the relevant search candidate point in a past search result, a search prediction time corresponding to each search candidate point, and a confidence interval upper limit corresponding to each search candidate point, and determining a search point based on a size of the area in a plurality of parameter spaces.

Abstract: Rule induction is used to produce human readable descriptions of patterns within a dataset. A rule induction algorithm or classifier is a type supervised machine learning classification algorithm. A rule induction classifier is trained, which involves using labelled examples in the dataset to produce a set of rules. Rather than using the rules/classifier to make predictions on new unlabeled samples, the training of the rule induction model outputs human-readable descriptions of patterns (rules) within the dataset that gave rise to the rules (rather than using the rules to predict new unlabeled samples). Parameters of the rule induction algorithm are tuned to favor simple and understandable rules, instead of only tuning for predictive accuracy. The learned set of rules are outputted during the training process in a human-friendly format.

Abstract: A method for operating a distributed neural network having a plurality of intelligent devices and a server includes: generating, by a first intelligent device of the plurality of intelligent devices, a first output using a first neural network model running on the first intelligent device and using a first input vector to the first neural network model; outputting, by the first intelligent device, the first output; receiving, by the first intelligent device, a gesture feedback on the first output from a user; determining, by the first intelligent device, a user rating of the first output from the gesture feedback; labeling, by the first intelligent device, the first input vector with a first label in accordance with the user rating; and training, by the first intelligent device, the first neural network model using the first input vector and the first label.

Abstract: We describe systems and methods for generating and training convolutional neural networks using biological sequences and relevance scores derived from structural, biochemical, population and evolutionary data. The convolutional neural networks take as input biological sequences and additional information and output molecular phenotypes. Biological sequences may include DNA, RNA and protein sequences. Molecular phenotypes may include protein-DNA interactions, protein-RNA interactions, protein-protein interactions, splicing patterns, polyadenylation patterns, and microRNA-RNA interactions, which may be described using numerical, categorical or ordinal attributes. Intermediate layers of the convolutional neural networks are weighted using relevance score sequences, for example, conservation tracks.

Abstract: An information processing apparatus includes an itemized reliability level calculation unit configured to calculate a first reliability level, wherein the first reliability level is a reliability level of classification target data, and a second reliability level, wherein the second reliability level is a reliability level of a label associated with the classification target data, a learning data reliability level calculation unit configured to calculate a learning data reliability level of learning data including the classification target data and the label based on the first reliability level and the second reliability level, and a classification model learning unit configured to formulate a classification model for giving a label to desired classification target data based on plural pieces of learning data and learning data reliability levels.

Abstract: Methods, systems, and computer program products for active explanation guided learning are provided herein. A computer-implemented method includes identifying a subset of training examples, from a set of training examples, based on at least one of (i) an uncertainty metric computed for each one of the training examples and (ii) an influence metric computed for each one of the training examples; outputting said subset of training examples to a user; obtaining, from the user, a user explanation for each training example in said subset of training examples, wherein each of the user explanations identifies at least one part of the corresponding training example; and training a machine learning model based at least in part on the user explanations, wherein said training comprises prioritizing the identified parts of the training examples in the subset.

Abstract: A first sample acquisition unit acquires a parameter sample from a prior distribution. A function execution unit acquires data from a distribution by supplying the sample to a function. A degree-of-similarity calculation unit calculates the degree of similarity between the data and correct data using a kernel function. A kernel mean construction unit constructs a kernel mean of a posterior distribution from the degree of similarity, the sample, and the kernel function. A second sample acquisition unit acquires a new parameter sample from the kernel mean and the kernel function. A sample evaluation unit determines whether the difference between new data obtained by supplying one sample selected from the new samples to the function and the correct data is less than a prescribed threshold value. When it is determined that the difference is less than the prescribed threshold value, the sample evaluation unit estimates the selected sample as a parameter.

Abstract: Method(s) and apparatus are provided for interfacing with a nervous system of a subject.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for training a student neural network. In one aspect, there is provided a method comprising: processing a training input using the student neural network to generate a student neural network output comprising a respective score for each of a plurality of classes; processing the training input using a brain emulation neural network to generate a brain emulation neural network output comprising a respective score for each of the plurality of classes; and adjusting current values of the student neural network parameters using gradients of an objective function that characterizes a similarity between: (i) the student neural network output for the training input, and (ii) the brain emulation neural network output for the training input.

Abstract: In some implementations, a method is provided. The method includes determining a physical topology of a network and monitoring network events based, at least in part, on control plane information received from one or more devices in the network. The method also includes monitoring the performance of each of a plurality of applications running on the network based, at least in part, on a set of application calls initiated by each application. When a drop in performance of an application is detected, the drop in performance is correlated with one or more of a plurality of detected network events to determine a cause of the drop in performance.

Abstract: Systems and methods for image recognition are provided. A style-transfer neural network is trained for each real image to obtain a trained style-transfer neural network. The texture or style features of the real images are transferred, via the trained style-transfer neural network, to a target image to generate styled images which are used for training an image-recognition machine learning model (e.g., a neural network). In some cases, the real images are clustered and representative style images are selected from the clusters.

Abstract: Ground truth data may be too sparse to supervise training of a machine-learned (ML) model enough to achieve an ML model with sufficient accuracy/recall. For example, in some cases, ground truth data may only be available for every third, tenth, or hundredth frame of raw data. Training an ML model to detect a velocity of an object when ground truth data for training is sparse may comprise training the ML model to predict a future position of the object based at least in part on image, radar, and/or lidar data (e.g., for which no ground truth may be available). The ML model may be altered based at least in part on a difference between ground truth data associated with a future time and the future position.

Abstract: Computing systems, devices, and associated methods of detecting application regression in a distributed computing system are disclosed herein. In one embodiment, a method includes receiving data representing telemetry records from one or more hosts of the distributed computing system. At least some of the telemetry records are exception records individually indicating an operation by a user application has failed during execution. The method also includes determining a failure rate of executing the operation by the user application while compensating for a workload of the user application in the distributed computing system. A comparison is performed between the determined failure rate and a threshold. Based on the performed comparison, a regression notification can be generated to indicate that application regression has occurred notwithstanding the workload of the user application in the distributed computing system.

Abstract: Techniques are disclosed that enable automating user interface input by generating a sequence of actions to perform a task utilizing a multi-agent reinforcement learning framework. Various implementations process an intent associated with received user interface input using a holistic reinforcement policy network to select a software reinforcement learning policy network. The sequence of actions can be generated by processing the intent, as well as a sequence of software client state data, using the selected software reinforcement learning policy network. The sequence of actions are utilized to control the software client corresponding to the selected software reinforcement learning policy network.

Abstract: The image analysis apparatus includes: a first analysis unit that analyzes image frames by using a first image analysis model; a second analysis unit that can analyze the image frames by using a second image analysis model an analysis accuracy of which is lower; a storage unit that stores therein analyzed frames, which are already analyzed by using the first and second image analysis models, in association with an evaluation value for evaluating a result of an analysis performed with the second image analysis model; an extraction unit that extracts an analyzed frame that satisfies an extraction condition based on the evaluation value; and an update unit that updates the second image analysis model by using a result of an analysis performed with the first image analysis model on the extracted analyzed frame, and a result of an analysis performed with the first image analysis model on a new frame.

Abstract: According to various embodiments, systems, methods, and mediums for operating an autonomous driving vehicles (ADV) are described. The embodiments use a number of machine learning models to extract features individually from audio data and visual data captured by sensors mounted on the ADV, and then to fuse these extracted features to create a concatenated feature vectors. The concatenated feature vector is provided to a multiplayer perceptron (MLP) as input to generate a detection result related to the presence of an emergency vehicle in the surrounding environment. The detection result can be used by the ADV to take appropriate actions to comply with the local traffic rules.

Abstract: Described is a system for adapting a deep convolutional neural network (CNN). A deep CNN is first trained on an annotated source image domain. The deep CNN is adapted to a new target image domain without requiring new annotations by determining domain agnostic features that map from the annotated source image domain and a target image domain to a joint latent space, and using the domain agnostic features to map the joint latent space to annotations for the target image domain.

Abstract: An electronic device includes a memory that stores input matrices A and B, a cache memory, and a processor. The processor generates a compiled representation that includes values for acquiring data from input matrix A when processing instances of input data through the neural network, the values including a base address in input matrix A for each thread from among a number of threads and relative offsets, the relative offsets being distances between elements of input matrix A to be processed by the threads. The processor then stores, in the local cache memory, the compiled representation including the base address for each thread and the relative offsets.

Abstract: A method for improving food-related personalized for a user including determining food-related preferences associated with a plurality of users to generate a user food preferences database; collecting dietary inputs from a subject matter expert (SME) at an SME interface associated with the user food preferences database; determining personalized food parameters for the user based on the user food-related preferences and the dietary inputs; receiving feedback associated with the personalized food parameters from the user; and updating the user food preferences database based on the feedback.

Abstract: A method includes applying, by a computer, supervised machine learning to classify a network device that is associated with a security event occurring in a computer system based on data representing features of the network device. The security event is associated with a potential security threat to the computer system, and the classification of the network device by the supervised machine learning is associated with a confidence. The technique includes, in response to the confidence being below a threshold, applying an active machine learning classifier to the data to learn a classification for the data and using the classification learned by the active machine learning classifier to adapt the supervised machine learning to recognize the classification.

Abstract: A data management system trains an analysis model with a machine learning process to understand the semantic meaning of queries received from users of the data management system. The machine learning process includes retrieving assistance documents that each include a query and an answer to the query. A training model analyzes each answer and generates first topic distribution data indicating, for each answer, how relevant each of a plurality of topics is to the answer. The queries are passed to the analysis model and the analysis model is trained to generate second topic distribution data that converges with the first topic distribution data based on analysis of the queries.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for processing point cloud data using spatio-temporal-interactive networks.

Abstract: A processor may monitor frequency data related to a user metric of a user during a measurement window. The user metric may relate to the user's use of a computer implemented environment. The processor may simplify the frequency data related to the user metric, resulting in a set of simplified frequency data. The processor may input the set of simplified frequency data into a neural network. The neural network may determine a likelihood of a negative user experience for the user. The processor may alter a parameter of the first user environment based on the likelihood.

Abstract: Methods, systems, and computer-readable media are disclosed herein that employ a contextually intelligent framework. In accordance with embodiments, a knowledge model having rules, axioms, and a domain ontology is evaluated to determine rules that are redundant to other rules and axioms, to determines those rules thresholds that may be refactored to generate composite rules and reduce the overall quantity of rules in the knowledge model, and to generate and add new concepts as axioms to the domain ontology as determined through refactoring. Methods, systems, and computer-readable media are disclosed herein that use the refactored and improved knowledge model to reconcile information currently stored in one system with information imported from a plurality of diverse systems, in order to generate recommendations that promote continuity of care in clinical settings.

Abstract: Embodiments can provide a method for identifying a behavioral pattern from simulated transaction data, including: simulating transaction data using a reinforcement learning model; identifying a behavioral pattern from the simulated transaction data; comparing the behavioral pattern with standard customer transaction data to determine whether the behavioral pattern is present in the standard customer transaction data. If the behavioral pattern is present in the standard customer transaction data, the behavioral pattern is applied in a model implemented on the cognitive system.

Abstract: A method provides learning with noisy labels. The method includes generating a first network of a machine learning model with a first set of parameter initial values, and generating a second network of the machine learning model with a second set of parameter initial values. First clean probabilities for samples in a training dataset are generated using the second network. A first labeled dataset and a first unlabeled dataset are generated from the training dataset based on the first clean probabilities. The first network is trained based on the first labeled dataset and first unlabeled dataset to update parameters of the first network.

Abstract: Embodiments provide computer systems, computer-executable methods and one or more non-transitory computer-readable media for offering one or more promotions to consumers using a promotion and marketing service. User input may be received from a first consumer interface associated with a first consumer, the user input including an interest indication relating to a first promotion. An association between the first consumer and a second consumer for sharing of promotions may be programmatically retrieved or generated. Based on the association, it may be determined whether to offer the first promotion to the second consumer based on one or more characteristics associated with the second consumer. Based on a determination that the first promotion should be offered to the second consumer, an indication may be outputted, the indication configured to cause an impression of the first promotion to be generated on a second consumer interface associated with the second consumer.

Abstract: There is provide an information processing device capable of reducing the time taken for selection of the learning setting, the information processing device including: a data acquisition unit configured to acquire a learning setting corresponding to information related to previous learning processing in which a degree of similarity with information related to learning processing specified by a user is higher than a predetermined degree of similarity as a learning setting to be recommended to the user; and a display control unit configured to control display corresponding to the learning setting to be recommended.

Abstract: Certain aspects of the present disclosure provide techniques for placing targeted messages in communications within a software application using machine learning models. An example method generally includes retrieving, from a repository, a data set of targeted messages. For each respective targeted message in the data set of targeted messages, an effectiveness score for a party associated with the respective targeted message, a distance score between the party associated with the respective targeted message and a host party, and a match score between the party associated with the respective targeted message and the host party. Based on the effectiveness score, the distance score, and the match score for each respective targeted message in the data set of targeted messages, a message is selected to be included in one or more communications by the host party. The one or more communications including the selected message are generated and output for transmission.

Abstract: At a machine learning service, a data structure generated during the training phase of a machine learning model, as well as an input records associated with a result of the model, are analyzed. A first informational data set pertaining to the result, which indicates an alternative result, is generated. The first informational data set is transmitted to a presentation device with a directive to display a visual representation of the data set. In response to an exploration request pertaining to the first informational data set, a second informational data set indicating one or more observations of a training data set used for the model is transmitted to the presentation device.