Abstract: A method and apparatus for log files of data compression are disclosed. The method comprising: classifying each of a plurality of lines in a plurality of the log files of data with at least two levels hierarchy clustering comprising identifying a plurality of strings repeated in the plurality of lines of the plurality of log files of data. Creating a table matching each of the plurality of strings to a unique value. Creating a vector encoding the unique value matched to each of the plurality of strings using the table. Assigning each of the encoded unique values in the vector, a security relevance score according to the classification of the plurality of lines; and selecting a subset of the encoded unique values such that the encoded unique values in the vector are filtered according to the security relevance score of each unique value.

Abstract: Techniques are disclosed for detecting adversarial attacks. A machine learning (ML) system processes the input into and output of a ML model using an adversarial detection module that does not include a direct external interface. The adversarial detection module includes a detection model that generates a score indicative of whether the input is adversarial using, e.g., a neural fingerprinting technique or a comparison of features extracted by a surrogate ML model to an expected feature distribution for the output of the ML model. In turn, the adversarial score is compared to a predefined threshold for raising an adversarial flag. Appropriate remedial measures, such as notifying a user, may be taken when the adversarial score satisfies the threshold and raises the adversarial flag.

Abstract: Mechanisms and methods are provided for establishing vectors indicating the presence, in a first vehicle and second vehicle, of a super-set of vehicle features present across a fleet of vehicles. The first vehicle may be a reference vehicle. A distance function of the vectors may be calculated in order to establish a similarity score indicating the degree of similarity between the designs of the two vehicles. If the second vehicle is sufficiently similar to the reference vehicle, a software update may be recommended and applied.

Abstract: Disclosed herein are systems and methods for detecting potential malicious attacks in vehicles operational environment using staged Machine Learning (ML), comprising creating a plurality of features vectors each comprising a plurality of features extracted from vehicle operational data generated by a plurality of devices deployed in one or more vehicles which is indicative of operation of the one or more vehicles, detecting, in real-time, a plurality of anomaly feature vectors using one or more unsupervised ML models applied to the plurality of feature vectors, identifying, in real-time, one or more potential cyberattack events using one or more supervised ML models applied to the plurality of anomaly feature vectors, and generating an alert indicative of the one or more potential cyberattack events.

Abstract: Mechanisms and methods are provided for establishing vectors indicating the presence, in a first vehicle and second vehicle, of a super-set of vehicle features present across a fleet of vehicles. The first vehicle may be a reference vehicle. A distance function of the vectors may be calculated in order to establish a similarity score indicating the degree of similarity between the designs of the two vehicles. If the second vehicle is sufficiently similar to the reference vehicle, a software update may be recommended and applied.

Abstract: Techniques are disclosed for detecting adversarial attacks. A machine learning (ML) system processes the input into and output of a ML model using an adversarial detection module that does not include a direct external interface. The adversarial detection module includes a detection model that generates a score indicative of whether the input is adversarial using, e.g., a neural fingerprinting technique or a comparison of features extracted by a surrogate ML model to an expected feature distribution for the output of the ML model. In turn, the adversarial score is compared to a predefined threshold for raising an adversarial flag. Appropriate remedial measures, such as notifying a user, may be taken when the adversarial score satisfies the threshold and raises the adversarial flag.