Abstract: A method is provided that includes: receiving a plurality of estimated position points, each estimated position point including a timestamp, where each estimated position point is an estimate of a position of a vehicle at a time respective timestamp; receiving on or more path events, where each of the one or more path events includes a timestamp and data from at least one sensor of the vehicle; generating a path from the plurality of estimated position points, where the estimated position points are arranged in order of ascending time represented by the respective timestamp; and interpolating between two of the estimated position points to determine a location corresponding to one of the one or more path events, where the timestamp of the one of the one or more path event corresponds to a time that is between the times represented by the timestamps of the two estimated position points.

Abstract: An approach is provided for compression of sparse data for machine learning or equivalent tasks. The approach involves, for instance, receiving data that is binned into a plurality of bins. The data, for instance, represents a spatial surface such as a geographic region. The approach also involves processing the data by applying a compression criterion to classify one or more bins of the plurality of bins as either data-containing bins or empty bins. The approach further involves establishing a space filling curve over the plurality of bins, wherein the space filling curve linearizes the plurality of bins according to a placement order. The approach further involves storing the data-containing bins of the plurality of bins in a compressed data structure based on the placement order of the space filling curve.

Abstract: Systems and methods are provided for training a machine learned model on a large number of devices, each device acquiring a local set of training data without sharing data sets across devices. The devices train the model on the respective device's set of training data. The devices communicate a parameter vector from the trained model asynchronously with a parameter server. The parameter server updates a master parameter vector and transmits the master parameter vector to the respective device.

Abstract: Systems, methods, and apparatuses are disclosed for predicting or estimating the value of a variable speed sign (VSS). A variable speed sign is identified. Probe data is collected at one or more vehicles in proximity to the variable speed sign. The speeds of the vehicles are included in or derived from the probe data. A statistical analysis is performed on the probe data. A speed limit value for the variable speed sign is determined based on the statistical analysis.

Abstract: Systems and methods are provided for managing machine learning processes within distributed and heterogeneous environments. The distributed and heterogeneous environments may include different types of devices that include different specifications, security, and privacy concerns. The devices participate in complex machine learning tasks while maintaining both privacy and autonomy. The systems and methods manage the lifecycle of how machine learning workloads are distributed.

Abstract: Systems and methods are provided for training a model on a large number of devices where, for example, each device acquires a local set of training data without sharing data sets across the devices. The devices train the model on the respective device's set of training data. The devices communicate a parameter vector from the trained model asynchronously with a parameter server. The parameter server updates a master parameter vector and transmits the master parameter vector to the respective device. The update rate of the devices is decoupled from the size of the data that is available to the devices and the computational power of the devices by over or under sampling the local training data.

Abstract: Systems and methods are provided for training a machine learned model on a large number of devices, each device acquiring a local set of training data without sharing data sets across devices. The devices train the model on the respective device's set of training data. The devices communicate a parameter vector from the trained model asynchronously with a parameter server. The parameter server updates a master parameter vector and transmits the master parameter vector to the respective device.

Abstract: Systems, methods, and apparatuses are disclosed for predicting or estimating the value of a variable speed sign (VSS). A variable speed sign is identified. Probe data is collected at one or more vehicles in proximity to the variable speed sign. The speeds of the vehicles are included in or derived from the probe data. A statistical analysis is performed on the probe data. A speed limit value for the variable speed sign is determined based on the statistical analysis.

Abstract: Systems, methods, and apparatuses are disclosed for predicting or estimating the value of a variable speed sign (VSS). A variable speed sign is identified. Probe data is collected at one or more vehicles in proximity to the variable speed sign. The speeds of the vehicles are included in or derived from the probe data. A statistical analysis is performed on the probe data. A speed limit value for the variable speed sign is determined based on the statistical analysis.

Abstract: An approach is provided for providing map independent location-based notifications. The location-based notification platform causes an initiation of at least one spatial event request for one or more location-based notifications. The at least one spatial event request specifies at least one bounded area in which at least one device is located or to arrive. The location-based notification platform in response to the at least one spatial event request, receives at least one spatial polygon. The at least one spatial polygon represents at least one geographic area that is determined based on the at least one bounded area, and the at least one spatial polygon is associated with the one or more location-based notifications or one or more triggers for the one or more location-based notifications. The location-based notification platform then causes an overlay of the at least one spatial polygon on at least one travel network traveled by the at least one device.

Abstract: A method is provided that includes: receiving a plurality of estimated position points, each estimated position point including a timestamp, where each estimated position point is an estimate of a position of a vehicle at a time respective timestamp; receiving on or more path events, where each of the one or more path events includes a timestamp and data from at least one sensor of the vehicle; generating a path from the plurality of estimated position points, where the estimated position points are arranged in order of ascending time represented by the respective timestamp; and interpolating between two of the estimated position points to determine a location corresponding to one of the one or more path events, where the timestamp of the one of the one or more path event corresponds to a time that is between the times represented by the timestamps of the two estimated position points.

Abstract: An approach is provided for providing map independent location-based notifications. The location-based notification platform causes an initiation of at least one spatial event request for one or more location-based notifications. The at least one spatial event request specifies at least one bounded area in which at least one device is located or to arrive. The location-based notification platform in response to the at least one spatial event request, receives at least one spatial polygon. The at least one spatial polygon represents at least one geographic area that is determined based on the at least one bounded area, and the at least one spatial polygon is associated with the one or more location-based notifications or one or more triggers for the one or more location-based notifications. The location-based notification platform then causes an overlay of the at least one spatial polygon on at least one travel network traveled by the at least one device.

Abstract: Systems, methods, and apparatuses are disclosed for predicting or estimating the value of a variable speed sign (VSS). A variable speed sign is identified. Probe data is collected at one or more vehicles in proximity to the variable speed sign. The speeds of the vehicles are included in or derived from the probe data. A statistical analysis is performed on the probe data. A speed limit value for the variable speed sign is determined based on the statistical analysis.