Abstract: Data is received that characterizes artefacts associated with each of a plurality of layers of a first machine learning model. Fingerprints are then generated for each of the artefacts in the layers of the first machine learning model. These generated fingerprints collectively form a model indicator for the first machine learning model. It is then determined whether the first machine learning model is derived from another machine learning model by performing a similarity analysis between the model indicator for the first machine learning model and model indicators generated for each of a plurality of reference machine learning models each comprising a respective set of fingerprints. Data characterizing the determination can be provided to a consuming application or process. Related apparatus, systems, techniques and articles are also described.

Abstract: A method and system for generating two or three dimensional computer graphics images using multisample antialiasing (MSAA) is provided, which enables memory bandwidth to be conserved. For each of one or more pixels it is determined whether all of a plurality of sample areas of that pixel are located within a particular primitive. For those pixels where it is determined that all the sample areas of that pixel are located within that primitive, a value is stored in a multisample memory for a smaller number of the sample areas of that pixel than the total number of the sample areas of that pixel and data is stored indicating that all the sample areas of that pixel are located within that primitive.

Abstract: A non-transitory computer readable medium includes a set of instructions that in operation cause a processor to determine at least one modelled parameter of a feature of interest in petroleum exploration. The instructions also cause a processor to assign a likelihood value to each modelled parameter of the at least one modelled parameter and to generate an initial chance map for each modelled parameter of the at least one modelled parameter. Further, the instructions cause a processor to assign a weighting factor for each modelled parameter of the at least one modelled parameter, and to combine the initial chance maps using the weighting factor for each modelled parameter of the at least one modelled parameter to generate a first simulation chance map.

Abstract: An analysis engine receives data characterizing a prompt for ingestion by a generative artificial intelligence (GenAI) model. An intermediate result of the GenAI model or a proxy of the GenAI model responsive to the prompt is obtained. The analysis engine, using a prompt injection classifier and the intermediate result, determines whether the prompt comprises or is indicative of malicious content or elicits malicious actions. Data characterizing the determination is provided to a consuming application or process. Related apparatus, systems, techniques and articles are also described.

Abstract: The disclosure includes methods, systems and apparatus for predictive management of efficient selecting and receiving of retail electric utility service to a facility for a period, by automated selecting of a retail utility service provider corresponding to a selected least cost path of predicted rate plan choices across the period, wherein costs of all possible, viable time-bounded predicted rate plan choices are determined for predicted consumer usage where a predicted market of retail rate formulas for the period are predicted in relation to at least one variable, such as weather.

Abstract: A system includes a processor and a memory. The memory includes instructions that are executable by the processor to access training data of a modern feature of interest from direct observations, remotely determined data, or a combination thereof. The instructions are also executable to compile parameter data from at least one model simulation that impacts the modern feature of interest. The instructions are executable to train a machine-learning model to generate a predictive model that matches the training data of the modern feature of interest using the compiled parameter data as input. Furthermore, the instructions are executable to predict a feature of interest in a past time period using the predictive model and at least one historical model simulation that impacts the feature of interest. Additionally, the instructions are executable to execute a processing operation for facilitating hydrocarbon exploration based on the predicted feature of interest from the predictive model.

Abstract: An electric hair grooming appliance includes a handle and a hair grooming device. The electric hair grooming appliance also includes a pivot assembly pivotably attaching the hair grooming device to the handle. The pivot assembly includes a spherical housing. The hair grooming device and the spherical housing are pivotable relative to the handle about at least two pivot axes extending through the spherical housing. The electric hair grooming appliance further includes a motor contained in the spherical housing. The motor is configured to drive the hair grooming device.

Abstract: A computer-implemented method for targeted narrative-structure representation of a network computing environment. The method may include: monitoring the network computing environment; obtaining, via the monitoring, data relating to an action performed in the network computing environment; selecting a first preference set from a plurality of preference sets, each of the preference sets corresponding to a respective one of a plurality of user roles and including a first structural narrative definition configured for the corresponding one of the user roles; generating a first natural language narrative describing the action based on the selected first preference set; and exposing or providing a representation of the first natural language narrative to an individual user assigned to a first user role of the user roles corresponding to the first preference set.

Abstract: An analysis engine receives data characterizing a prompt for ingestion by a generative artificial intelligence (GenAI) model. An intermediate result of the GenAI model or a proxy of the GenAI model responsive to the prompt is obtained. The analysis engine, using a prompt injection classifier and the intermediate result, determines whether the prompt comprises or is indicative of malicious content or elicits malicious actions. Data characterizing the determination is provided to a consuming application or process. Related apparatus, systems, techniques and articles are also described.

Abstract: A method comprising obtaining a thickness for each of one or more sediment packages of a subsurface formation. The method comprises generating a thickness profile of each of the one or more sediment packages based on the thickness. The method comprises obtaining one or more properties of each of the one or more sediment packages based on the thickness profile. The method comprises generating, via a learning machine, one or more sediment package classifications based on the one or more properties. The method comprises and performing a subsurface operation based on the one or more sediment package classifications.

Abstract: A computer-implemented method for automated damage assessment of electric grid infrastructure for post-storm recovery includes a virtual model system including a three-dimensional (3D) virtual model of physical grid infrastructure with assets, using a lidar imaging data point cloud collected in the field to characterize the physical asset condition, performing a simulation algorithm including a damage simulation and a lidar imaging data simulation, and executing a trained classifier to identify the damaged or undamaged condition of the asset using the field collected lidar imaging data point cloud, where the classifier is trained with simulated collected lidar imaging data of undamaged and damaged assets.

Abstract: Disclosed is a consumable for use with apparatus for heating aerosolisable material to volatilise at least one component of the aerosolisable material. The consumable comprises an outer tube, an inner member inside the outer tube, and at least one support that supports the inner member relative to the outer tube so that at least one air gap exists between the inner member and the outer tube. At least one of the inner member, the outer tube and the support comprises aerosolisable material that is heatable to generate aerosol in the air gap. The consumable has at least one outlet for permitting the aerosol to pass out of consumable from the air gap.

Abstract: The disclosure relates to compositions comprising a first structure of the formula: or a salt thereof, wherein R1, R2, R3, Ar1, and n are defined herein. Materials, devices, and methods of using such compositions are also described.

Abstract: In one respect, there is provided a system for training a neural network adapted for classifying one or more scripts. The system may include at least one processor and at least one memory. The memory may include program code which when executed by the at least one memory provides operations including: receiving a disassembled binary file that includes a plurality of instructions; processing the disassembled binary file with a convolutional neural network configured to detect a presence of one or more sequences of instructions amongst the plurality of instructions and determine a classification for the disassembled binary file based at least in part on the presence of the one or more sequences of instructions; and providing, as an output, the classification of the disassembled binary file. Related computer-implemented methods are also disclosed.

Abstract: Data is received that characterizes artefacts associated with each of a plurality of layers of a first machine learning model. Fingerprints are then generated for each of the artefacts in the layers of the first machine learning model. These generated fingerprints collectively form a model indicator for the first machine learning model. It is then determined whether the first machine learning model is derived from another machine learning model by performing a similarity analysis between the model indicator for the first machine learning model and model indicators generated for each of a plurality of reference machine learning models each comprising a respective set of fingerprints. Data characterizing the determination can be provided to a consuming application or process. Related apparatus, systems, techniques and articles are also described.

Abstract: A system can determine a heterogeneity and a score for a reservoir for optimizing a drilling location. The system can receive a wireline log associated with a well that is positioned in a subterranean formation that includes a reservoir. The system can determine, using the wireline log, at least one statistical parameter for an interval of the well. The system can determine, using the at least one statistical parameter, a vertical heterogeneity of the reservoir. The system can determine, using the vertical heterogeneity, a score associated with the reservoir. The score can indicate an extraction difficulty and a carbon intensity of the reservoir. The system can output the score for optimizing a drilling location.

Abstract: A desalination system including: a partitioned container having upstream and downstream compartments divided by a movable partition, a first inlet port at an upstream end, and a second inlet port and an outlet port at the downstream end; a membrane container housing a cross-flow semipermeable membrane dividing the membrane container into saline and desalinated compartments, the saline compartment including first and second cross-flow ports, the desalinated compartment including a desalinated water outlet port; a feed pump connected to the first inlet port; a recharge pump having an inlet connected to the second cross-flow port and an outlet connected to the second inlet port; a main valve connected between the outlet port and the first cross-flow port; a bypass valve connecting the inlet port to the second cross-flow port and the recharge pump inlet; and a purge valve connecting the first cross-flow port and the main valve to a purge port.

Abstract: Data is received that characterizes artefacts associated with each of a plurality of layers of a first machine learning model. Fingerprints are generated corresponding to each of the artefacts in each layer. The generated fingerprints can collectively form a model indicator for the first machine learning model. A second machine learning model then determines, based on the generated fingerprints, whether the first machine learning model is derived from another machine learning model. Data provided this characterization can be provided to a consuming application or process. This second machine learning model can be trained model with historical fingerprints having a known provenance classification. Related apparatus, systems, techniques and articles are also described.

Abstract: A system is provided for classifying an instruction sequence with a machine learning model. The system may include at least one processor and at least one memory. The memory may include program code that provides operations when executed by the at least one processor. The operations may include: processing an instruction sequence with a trained machine learning model configured to detect one or more interdependencies amongst a plurality of tokens in the instruction sequence and determine a classification for the instruction sequence based on the one or more interdependencies amongst the plurality of tokens; and providing, as an output, the classification of the instruction sequence. Related methods and articles of manufacture, including computer program products, are also provided.

Abstract: Methods, non-transitory computer readable media, and computing devices are disclosed that obtains and stores a mapping of status values to process-blocks within stages associated with a linear process performed on an enterprise application network. A work item status in change records of the database is correlated with the status values based on the mapping to identify one of the stages for work items. Event records are then generated based on the correlation, activity records of the database, and another correlation of a unique identifier of one of the work items in each of the change and activity records. A generated value stream map (VSM) is then output for display that comprises a graphical timeline of activity associated with the linear process and metrics generated based on the event records and comprising real-time metrics for the process-blocks or stages and aggregate metrics for the linear process.