Abstract: Random cut trees are generated with respective to respective samples of a baseline set of data records of a data set for which outlier detection is to be performed. To construct a particular random cut tree, an iterative splitting technique is used, in which the attribute along which a given set of data records is split is selected based on its value range. With respect to a newly-received data record of the stream, an outlier score is determined based at least partly on a potential insertion location of a node representing the data record in a particular random cut tree, without necessarily modifying the random cut tree.

Abstract: Systems and methods provide for routing an autonomous vehicles based on an assessment of its limitations. The autonomous vehicle can avoid problematic routes by applying a set of rules to a map, where the set of rules are associated with a feature that causes a failure of the autonomous vehicle due to an autonomous vehicle limitation. Based on the application of the set of rules, a portion of a route on the map that is associated with the feature can be identified, and an identifier can be applied to the identified portion of the route associated with the feature on a routing map. The portion of the route can be omitted from a routable graph applied to the routing map, such that the route for the autonomous vehicle is generated in accordance with the routable graph.

Abstract: Systems and methods provide for routing autonomous vehicles by simulating an autonomous vehicle traversing a route on a map representing a physical area. A router flag can be generated at a specific location on the map. A router flag rule that defines a factor that caused a failure of the autonomous vehicle can then be created at the location, and the autonomous vehicle can be simulated traversing the to identify at least one other area on the map where the autonomous vehicle is likely to fail. An identifier can be applied to the other area on the map where the autonomous vehicle is likely to fail according to the router flag rule. The other area with the applied identifier can be omitted from a routable graph applied to the routing map, such that the route for the autonomous vehicle is generated in accordance with the routable graph.

Abstract: Random cut trees are generated with respective to respective samples of a baseline set of data records of a data set for which outlier detection is to be performed. To construct a particular random cut tree, an iterative splitting technique is used, in which the attribute along which a given set of data records is split is selected based on its value range. With respect to a newly-received data record of the stream, an outlier score is determined based at least partly on a potential insertion location of a node representing the data record in a particular random cut tree, without necessarily modifying the random cut tree.

Abstract: Random cut trees are generated with respective to respective samples of a baseline set of data records of a data set for which outlier detection is to be performed. To construct a particular random cut tree, an iterative splitting technique is used, in which the attribute along which a given set of data records is split is selected based on its value range. With respect to a newly-received data record of the stream, an outlier score is determined based at least partly on a potential insertion location of a node representing the data record in a particular random cut tree, without necessarily modifying the random cut tree.

Abstract: Systems and methods provide for routing autonomous vehicles by simulating an autonomous vehicle traversing a route on a map representing a physical area. A router flag can be generated at a specific location on the map. A router flag rule that defines a factor that caused a failure of the autonomous vehicle can then be created at the location, and the autonomous vehicle can be simulated traversing the to identify at least one other area on the map where the autonomous vehicle is likely to fail. An identifier can be applied to the other area on the map where the autonomous vehicle is likely to fail according to the router flag rule. The other area with the applied identifier can be omitted from a routable graph applied to the routing map, such that the route for the autonomous vehicle is generated in accordance with the routable graph.

Abstract: Systems and methods provide for routing an autonomous vehicles based on an assessment of its limitations. The autonomous vehicle can avoid problematic routes by applying a set of rules to a map, where the set of rules are associated with a feature that causes a failure of the autonomous vehicle due to an autonomous vehicle limitation. Based on the application of the set of rules, a portion of a route on the map that is associated with the feature can be identified, and an identifier can be applied to the identified portion of the route associated with the feature on a routing map. The portion of the route can be omitted from a routable graph applied to the routing map, such that the route for the autonomous vehicle is generated in accordance with the routable graph.

Abstract: An indication of a data source to be used to train a linear prediction model is obtained. The model is to generate predictions using respective parameters assigned to a plurality of features derived from observation records of the data source. The parameter values are stored in a parameter vector. During a particular learning iteration of the training phase of the model, one or more features for which parameters are to be added to the parameter vector are identified. In response to a triggering condition, parameters for one or more features are removed from the parameter vector based on an analysis of relative contributions of the features represented in the parameter vector to the model's predictions. After the parameters are removed, at least one parameter is added to the parameter vector.

Abstract: Random cut trees are generated with respective to respective samples of a baseline set of data records of a data set for which outlier detection is to be performed. To construct a particular random cut tree, an iterative splitting technique is used, in which the attribute along which a given set of data records is split is selected based on its value range. With respect to a newly-received data record of the stream, an outlier score is determined based at least partly on a potential insertion location of a node representing the data record in a particular random cut tree, without necessarily modifying the random cut tree.

Abstract: An indication of a data source to be used to train a linear prediction model is obtained. The model is to generate predictions using respective parameters assigned to a plurality of features derived from observation records of the data source. The parameter values are stored in a parameter vector. During a particular learning iteration of the training phase of the model, one or more features for which parameters are to be added to the parameter vector are identified. In response to a triggering condition, parameters for one or more features are removed from the parameter vector based on an analysis of relative contributions of the features represented in the parameter vector to the model's predictions. After the parameters are removed, at least one parameter is added to the parameter vector.