Abstract: Herein are timeseries preprocessing, model selection, and hyperparameter tuning techniques for forecasting development based on temporal statistics of a timeseries and a single feed-forward pass through a machine learning (ML) pipeline. In an embodiment, a computer hosts and operates the ML pipeline that automatically measures temporal statistic(s) of a timeseries. ML algorithm selection, cross validation, and hyperparameters tuning is based on the temporal statistics of the timeseries. The result from the ML pipeline is a rigorously trained and production ready ML model that is validated to have increased accuracy for multiple prediction horizons. Based on the temporal statistics, efficiency is achieved by asymmetry of investment of computer resources in the tuning and training of the most promising ML algorithm(s). Compared to other approaches, this ML pipeline produces a more accurate ML model for a given amount of computer resources and consumes fewer computer resources to achieve a given accuracy.

Abstract: For machine learning of timeseries forecasting, here is discovery of multiple seasonalities by autoregression. A multivariate timeseries contains a variable that has a first seasonality that has a first period and a second seasonality that has a second period that is longer than the first period. Many local maxima of an autocorrelation of the variable in the timeseries are selected. For each local maximum as a distinct lag, a candidate feature that lags the variable based on the distinct lag is inserted into the timeseries. Feature selection selects a minimal subset of the candidate features. Based on the minimal subset of features, a feature vector that represents a point in the timeseries is generated. From the feature vector, a future value for the variable is predicted.

Abstract: A Contrastive Forecasting Explanation (CFE) tool and technique provides a model-agnostic approach to forecasting explanation. The CFE tool uses an ML-based surrogate forecaster as a surrogate model. The surrogate forecaster includes a time series preprocessor, a simple concept generator, and an ML forecaster. The subsequent interpretation of the predictions of the time series forecaster is based on the behavior of the surrogate forecaster. The CFE tool interprets time series forecasts by identifying the specific temporal concepts impacting predictions and thus generates clear and reliable explanations regardless of model type. The simple concepts and predictions generated by the surrogate model are input into a perturbation-based explainer to produce feature attributions from the surrogate model. An attribution postprocessor aggregates the attributions into more coherent concepts to present a coherent, concise, and interpretable explanation.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for sanitizing artificial intelligence prompts. One of the methods includes receiving a message a) for an external system and b) that comprises two or more phrases; for at least one phrase from the two or more phrases: determining a context of the phrase in the message; determining, using the context, whether modification of the phrase will likely maintain an intent of the message; determining whether to permit unedited transmission of the message to the external system using a result of at least one of one or more determinations whether modification of the phrase will likely maintain the intent of the message; and performing one or more actions using a result of the determination whether to permit unedited transmission of the message to the external system.

Abstract: In an embodiment, a computer combines first original hyperparameters and second original hyperparameters into combined hyperparameters. In each iteration of a binary search that selects hyperparameters, these are selected: a) important hyperparameters from the combined hyperparameters and b) based on an estimated complexity decrease by including only important hyperparameters as compared to the combined hyperparameters, which only one boundary of the binary search to adjust. For the important hyperparameters of a last iteration of the binary search that selects hyperparameters, a pruned value range of a particular hyperparameter is generated based on a first original value range of the particular hyperparameter for the first original hyperparameters and a second original value range of the same particular hyperparameter for the second original hyperparameters.

Abstract: In another aspect, a computerized method comprising: generating a strong envelope around a user's private key using a Shamir's Secret Sharing technique on a plurality of simple passwords to generate a single complex secret through cryptographic intermingling of the multiple passwords; and intermingling of a subset of originally configured passwords to reconstruct the complex secret for decrypting the envelope around user's private key.

Abstract: Herein are timeseries preprocessing, model selection, and hyperparameter tuning techniques for forecasting development based on temporal statistics of a timeseries and a single feed-forward pass through a machine learning (ML) pipeline. In an embodiment, a computer hosts and operates the ML pipeline that automatically measures temporal statistic(s) of a timeseries. ML algorithm selection, cross validation, and hyperparameters tuning is based on the temporal statistics of the timeseries. The result from the ML pipeline is a rigorously trained and production ready ML model that is validated to have increased accuracy for multiple prediction horizons. Based on the temporal statistics, efficiency is achieved by asymmetry of investment of computer resources in the tuning and training of the most promising ML algorithm(s). Compared to other approaches, this ML pipeline produces a more accurate ML model for a given amount of computer resources and consumes fewer computer resources to achieve a given accuracy.

Abstract: Interpretive meta-instructions are used, at a connected vehicle access system (CVAS) device, to implement various RKE protocols by: receiving an RKE-protocol-specific meta-instruction that has been generated by a back-end server; generating an RKE telegram based on the RKE-protocol-specific meta-instruction; and transmitting the RKE telegram to a vehicle. The meta-instruction may include: a device configuration that specifies a set of RF parameters of a CVAS-device transceiver; a template telegram that contains a first set of data fields with data values from the back-end server and that contains a second set of data fields for which data values will be generated and inserted at the CVAS device; and localization-processing commands containing instructions for modifying the second set of data fields of the template telegram.

Abstract: Interpretive meta-instructions are used, at a connected vehicle access system (CVAS) device, to implement various RKE protocols by: receiving an RKE-protocol-specific meta-instruction that has been generated by a back-end server; generating an RKE telegram based on the RKE-protocol-specific meta-instruction; and transmitting the RKE telegram to a vehicle. The meta-instruction may include: a device configuration that specifies a set of RF parameters of a CVAS-device transceiver; a template telegram that contains a first set of data fields with data values from the back-end server and that contains a second set of data fields for which data values will be generated and inserted at the CVAS device; and localization-processing commands containing instructions for modifying the second set of data fields of the template telegram.

Abstract: A Bluetooth enabled Smartphone may be used for both access control and start authorization in a secure and safe way, and embodiments are backward-compatible with conventional vehicle access and start systems. A smart phone acts as an intermediary authorization device to a code generator which effectively resembles a car key that is installed in a vehicle. A Bluetooth transceiver and the code generator—and, optionally, for the retrofit solution, an RF/LF transceiver—are added to the vehicle. The Bluetooth transceiver communicates with the smart phone. The code generator communicates with electronic control units in the vehicle that control access, immobilization, and engine start. The communication may happen via a wired connection or, in the case of the retrofit solution, via an RF/LF transceiver that mimics an additional car key programmed to the vehicle.

Abstract: A method to learn and then pair with a pre-installed access control system of a vehicle is discussed. Communication is exchanged between the access control system and a backend cloud-based system. Required data of the access control system including its particular authentication code is extracted by a learning device. A vehicle matching data is sent to the backend cloud-based system and the vehicle is registered with the backend cloud-based system. The learning device is registered to the access control system in accordance with learning procedures implemented in the vehicle as remote entry key. The learning device is coupled to a Radio Frequency signal transmitter that has Application-Specific Integrated Circuits to generate stable RF signals at multiple frequency wavelengths.

Abstract: A Bluetooth enabled Smartphone may be used for both access control and start authorization in a secure and safe way, and embodiments are backward-compatible with conventional vehicle access and start systems. A smart phone acts as an intermediary authorization device to a code generator which effectively resembles a car key that is installed in a vehicle. A Bluetooth transceiver and the code generator—and, optionally, for the retrofit solution, an RF/LF transceiver—are added to the vehicle. The Bluetooth transceiver communicates with the smart phone. The code generator communicates with electronic control units in the vehicle that control access, immobilization, and engine start. The communication may happen via a wired connection or, in the case of the retrofit solution, via an RF/LF transceiver that mimics an additional car key programmed to the vehicle.

Abstract: A method to learn and then pair with a pre-installed access control system of a vehicle is discussed. Communication is exchanged between the access control system and a backend cloud-based system. Required data of the access control system including its particular authentication code is extracted by a learning device. A vehicle matching data is sent to the backend cloud-based system and the vehicle is registered with the backend cloud-based system. The learning device is registered to the access control system in accordance with learning procedures implemented in the vehicle as remote entry key. The learning device is coupled to a Radio Frequency signal transmitter that has Application-Specific Integrated Circuits to generate stable RF signals at multiple frequency wavelengths.

Abstract: A key fob simulator for sending actuation command to a vehicle is discussed. The key fob has memory buffers, processors, and a transceiver that uses wireless communications to communicate with a backend cloud-based system. A RF transmitter of the key fob can transmit RF signals to Remote Keyless Entry (RKE) module of the vehicle. A mapping module includes a map-calculating circuit to calculate map coordinates of the key fob. A security module can receive a rolling security key of the RKE module of the vehicle. The key fob includes buttons that can be pushed by a user of the key fob to generate actuation commands by the security module. Using the RF transmitter, an actuation command and the rolling security key can be sent from security module to the RKE module of the vehicle. The RKE module then executes the actuation command after validating the rolling security key.

Abstract: A method to learn and then pair with a pre-installed access control system of a vehicle is discussed. Communication is exchanged between the access control system and a backend cloud-based system. Required data of the access control system including its particular authentication code is extracted by a learning device. A vehicle matching data is sent to the backend cloud-based system and the vehicle is registered with the backend cloud-based system. The learning device is registered to the access control system in accordance with learning procedures implemented in the vehicle as remote entry key. The learning device is coupled to a Radio Frequency signal transmitter that has Application-Specific Integrated Circuits to generate stable RF signals at multiple frequency wavelengths.

Abstract: A key fob simulator for sending actuation command to a vehicle is discussed. The key fob has memory buffers, processors, and a transceiver that uses wireless communications to communicate with a backend cloud-based system. A RF transmitter of the key fob can transmit RF signals to Remote Keyless Entry (RKE) module of the vehicle. A mapping module includes a map-calculating circuit to calculate map coordinates of the key fob. A security module can receive a rolling security key of the RKE module of the vehicle. The key fob includes buttons that can be pushed by a user of the key fob to generate actuation commands by the security module. Using the RF transmitter, an actuation command and the rolling security key can be sent from security module to the RKE module of the vehicle. The RKE module then executes the actuation command after validating the rolling security key.

Abstract: A method to learn and then pair with a pre-installed access control system of a vehicle is discussed. Communication is exchanged between the access control system and a backend cloud-based system. Required data of the access control system including its particular authentication code is extracted by a learning device. A vehicle matching data is sent to the backend cloud-based system and the vehicle is registered with the backend cloud-based system. The learning device is registered to the access control system in accordance with learning procedures implemented in the vehicle as remote entry key. The learning device is coupled to a Radio Frequency signal transmitter that has Application-Specific Integrated Circuits to generate stable RF signals at multiple frequency wavelengths.

Abstract: A cloud based package transfer system is discussed that has a cloud based package-exchange-service hosted on a cloud based provider site. The cloud based package-exchange-service supplies servers and databases of retail websites with information regarding services for package exchange available to customers and their associated vehicles. The information enables a retail website to present a checkbox, the presented checkbox enables a customer at a checkout point of the purchase from the retail website to select an alternative package delivery option of delivering the purchased products to a target vehicle of the customer. The package-exchange-service receives purchase information from the retail websites and use purchase information to create database records for completed delivery operations.

Abstract: A method and system of assigning paths through an interconnection network (100) consisting of a plurality of switching elements (102:165) and a plurality of links coupling the switching elements (102:165) are provided. Physical restrictions of the interconnection network (100) are used to arrive at a logical representation of an architecture of the interconnection network (100). Traffic patterns of the interconnection network (100) are determined to balance the data traffic through the links coupling the switching elements (102:165). The logical representation and traffic patterns of the interconnection network (100) are used to setup virtual channel identifiers that determine paths through the switching elements and links so that data traffic is more evenly distributed through the interconnection network (100).

Abstract: A Content Addressable Memory (CAM) that downloads and banks a group of filters. Each filter has an associated filter-index, which is mapped in a mapping table to an available CAM address for storing the associated filter. Each succeeding filter in the group of filters is stored in a higher address in the CAM. The plurality of CAM addresses may not all be contiguous, in which case the mapping table maps each succeeding filter-index to the next higher available CAM address without regard to contiguity of the addresses.