Abstract: In one embodiment, a method of training dynamic models for autonomous driving vehicles includes the operations of receiving a first set of training data from a training data source, the first set of training data representing driving statistics for a first set of features; training a dynamic model based on the first set of training data for the first set of features; determining a second set of features as a subset of the first set of features based on evaluating the dynamic model, each of the second set of features representing a feature whose performance score is below a predetermined threshold. The method further includes the following operations for each of the second set of features: retrieving a second set of training data associated with the corresponding feature of the second set of features, and retraining the dynamic model using the second set of training data.

Abstract: Provided is multi-layer neural network technique that includes: calculating, from an input and using a first one or more layers of a plurality of layers of a neural network, a first intermediate output; reducing a size of one or more dimensions of the first intermediate output; calculating, from the first intermediate output and using a second one or more layers of the neural network, a second intermediate output (the second one or more layers including one or more ultra-low precision layers); reducing a size of one or more dimensions of the second intermediate output; combining a plurality of reduced intermediate outputs (including the reduced first intermediate output and the reduced second intermediate output) to derive a combined intermediate output; and calculating, using the combined intermediate output and one or more higher-precision layers of the plurality of layers, a neural network output.

Abstract: Machine learning techniques can be used to train a classifier, in some embodiments, to accurately detect similarities between different records of user activity for a same user. When more recent data is received, newer data can be analyzed by selectively removing particular sub-groups of data to see if there is any particular data that accounts for a large difference (e.g. when run through a classifier that has been trained to produce similar results for known activity data from a same user). If a sub-group of data is identified as being significantly different from other user data, this may indicate an account breach. Advanced machine learning techniques described herein may be applicable to a variety of different environments.

Abstract: Systems and methods for spatial graph convolutions in accordance with embodiments of the invention are illustrated. One embodiment includes a method for predicting characteristics for molecules, wherein the method includes performing a first set of graph convolutions with a spatial graph representation of a set of molecules, wherein the first set of graph convolutions are based on bonds between the set of molecules, performing a second set of graph convolutions with the spatial graph representation, wherein the second set of graph convolutions are based on at least a distance between each atom and other atoms of the set of molecules, performing a graph gather with the spatial graph representation to produce a feature vector, and predicting a set of one or more characteristics for the set of molecules based on the feature vector.

Abstract: A method comprising identifying a set of target features for a plurality of data instances of an input data collection; determining feature values for the set of target features for the plurality of data instances; identifying a plurality of outlier data instances based on the determined feature values; identifying a plurality of noisy data instances from the outlier data instances based on feature values of the plurality of noisy data instances, wherein a noisy data instance is identified based on a determination that noise is present in noisy data instance; and providing an indication of the plurality of noisy data instances.

Abstract: A computerized method to store aggregate information handling system interaction telemetry data representing levels of operational activity reported for a user of an information handling system in a monitoring system data repository memory device and for receiving aggregate information handling system interaction telemetry data for a plurality of other users crowd-sourced from a population of information handling systems accessed by a plurality of other users. An interaction signature platform may apply a supervised learning model algorithm to the aggregate information handling system interaction telemetry data for the user in comparison to the aggregate information handling system interaction telemetry data for the plurality of other users to determine at least one indirect identifier of the interaction telemetry data.

Abstract: From each of in-vehicle units in vehicles, a travel behavior data indicating a travel behavior of the vehicle is received and recorded in a travel history database. The travel behavior data of the vehicles are read from the travel history database; an avoidance action arising in each of the vehicles is detected based on the read travel behavior data. From the detected avoidance actions of the vehicles, a related avoidance action group being avoidance actions arising due to an identical object at different positions at different points of times in different target vehicles is extracted based on an occurrence position and an occurrence point of time of each detected avoidance action. Information on position change of the identical object as a cause of the extracted related avoidance action group is recorded as an alert target data.

Abstract: Provided is a method and system for analyzing data in a run model at an edge device in a network environment. The method includes acquiring, at the edge device, data from the cloud environment, running a predetermined run model associated with the edge device and performing a first determination process by determining whether data analysis result from the run model performed is greater than an acceptance threshold. When it is determined that the data analysis result is less than the acceptance threshold, the method further performs a second determination process by determining whether the data analysis result is greater than a consideration threshold. If greater than the consideration threshold, the data is stored as acquired data to be further considered, and transferred to a cloud server.

Abstract: Embodiments of the present invention are directed to a computer-implemented method for multiscale representation of input data. A non-limiting example of the computer-implemented method includes a processor receiving an original input. The processor downsamples the original input into a downscaled input. The processor runs a first convolutional neural network (“CNN”) on the downscaled input. The processor runs a second CNN on the original input, where the second CNN has fewer layers than the first CNN. The processor merges the output of the first CNN with the output of the second CNN and provides a result following the merging of the outputs.

Abstract: A method and system are provided for combining models. The method includes forming, by a computer having a processor and a memory, model pairs from a model ensemble that includes a plurality of models. The method further includes comparing the model pairs based on sets of output results produced by the model pairs to provide comparison results. The method also includes constructing, by the computer, a combination model from at least one of the model pairs based on the comparison results. The comparing step is performed using user-generated set-based feedback.

Abstract: A domain adaptation module is used to optimize a first domain derived from a second domain using respective outputs from respective parallel hidden layers of the domains.

Abstract: Implementations provide for use of spherical random features for polynomial kernels and large-scale learning. An example method includes receiving a polynomial kernel, approximating the polynomial kernel by generating a nonlinear randomized feature map, and storing the nonlinear feature map. Generating the nonlinear randomized feature map includes determining optimal coefficient values and standard deviation values for the polynomial kernel, determining an optimal probability distribution of vector values for the polynomial kernel based on a sum of Gaussian kernels that use the optimal coefficient values, selecting a sample of the vectors, and determining the nonlinear randomized feature map using the sampled vectors.

Abstract: A system and method for predictively following up with a user to improve medication adherence. The system includes a medication adherence monitoring apparatus for determining whether a user has taken a medication at a predetermined medication administration time, and a processor for categorizing each determination of whether a user has taken the medication at a predetermined time across a plurality of different dimensions, combining the plurality of different dimensions in a plurality of different combinations to generate a patient adherence score across each of the plurality of different combinations, and ranking a user in accordance with each of the plurality of different combinations. The system further includes a communication apparatus for contacting a user to encourage medication adherence in accordance with at least the ranking of the user in accordance with one or more of the plurality of different combinations.

Abstract: Techniques facilitating estimating and visualizing entity to agent collaboration to facilitate automated plan generation are provided. In one example, a computer-implemented method comprises generating, by a device operatively coupled to a processor, a plan based on receiving first input data associated with an instance model. The computer-implemented method also comprises generating, by the device, a revised plan based on receiving second input data, associated with a revised instance model, from an entity. Furthermore, the computer-implemented method comprises, tracking, by the device, a contribution of the entity as a function of a modification from the instance model to the revised instance model.

Abstract: A cognitive assignment engine (CAE) system attempts to infer semantic meaning from textual content of an incoming message in order to use the inferred meaning to assign the message to an appropriate responder. If the message contains insufficient textual content, the system identifies ontological structures comprised by the message's graphical content and classifies each structure as a function of the structure's location within the graphical content or of an intrinsic characteristic of the structure. The system then generates a message identifier by performing a computation on these classifications and uses the identifier to retrieve a previously stored graphical template that comprises ontological structures similar to those of the incoming message. The system associates the incoming message with a semantic meaning previously associated with the template, enabling the system to classify the message and to assign the message to the correct responder.

Abstract: A device can receive information associated with content that is to be provided to a first set of user devices. The device can receive information associated with a set of rules that identifies a set of conditions for providing the content to the first set of user devices. The device can receive network information associated with a second set of user devices. The device can determine that at least one condition, of the set of conditions identified in the set of rules, is satisfied based on the network information associated with the second set of user devices. The device can determine an action, associated with the content, to be performed based on determining that the at least one condition is satisfied, and can perform the action.

Abstract: A machine learning management apparatus identifies a maximum prediction performance score amongst a plurality of prediction performance scores corresponding to a plurality of models generated by executing each of a plurality of machine learning algorithms. As for a first machine learning algorithm having generated a model corresponding to the maximum prediction performance score, the machine learning management apparatus determines a first training dataset size to be used when the first machine learning algorithm is executed next time based on the maximum prediction performance score, first estimated prediction performance scores, and first estimated runtimes.

Abstract: An example operation may include one or more of generating, by a training participant client comprising a training dataset, a plurality of transaction proposals that each correspond to a training iteration for machine learning model training related to stochastic gradient descent, the machine learning model training comprising a plurality of training iterations, the transaction proposals comprising a gradient calculation performed by the training participant client, a batch from the private dataset, a loss function, and an original model parameter, receiving, by one or more endorser nodes of peers of a blockchain network, the plurality of transaction proposals, and evaluating each transaction proposal.

Abstract: Disclosed is an approach for performing auto-classification of documents. A machine learning framework is provided to analyze the document, where labels associated with certain documents can be propagated to other documents.

Abstract: A method for estimating a brain activity response following a stimulus of a person comprises the steps: providing a usage data set of the person from a personal device used by said person, wherein at least one usage attribute is associated to said usage data set, wherein attribute data is associated to each of the at least one usage attribute, providing a computational inference model, generated from a plurality of brain activity data sets and a plurality of usage data sets, wherein each brain activity data set comprises data derived from a brain activity response following a sensory stimulus, submitting the attribute data of each of the at least one usage attributes to said computational inference model, estimating a brain activity response following a sensory stimulus of said person by evaluating said computational inference model for the submitted attribute data. The method is useful to determine, for example the influence of intensive touch pad usage (of a smartphone) on somatosensory evoked potentials.