Provided are methods and systems for automatically generating input grammars for grammar-based fuzzing by utilizing machine-learning techniques and sample inputs. Neural-network-based statistical learning techniques are used for the automatic generation of input grammars. Recurrent neural networks are used for learning a statistical input model that is also generative in that the model is used to generate new inputs based on the probability distribution of the learnt model.

Systems and methods for generating motion forecast data for actors with respect to an autonomous vehicle and training a machine learned model for the same are disclosed. The computing system can include an object detection model and a graph neural network including a plurality of nodes and a plurality of edges. The computing system can be configured to input sensor data into the object detection model; receive object detection data describing the location of the plurality of the actors relative to the autonomous vehicle as an output of the object detection model; input the object detection data into the graph neural network; iteratively update a plurality of node states respectively associated with the plurality of nodes; and receive, as an output of the graph neural network, the motion forecast data with respect to the plurality of actors.

A method of processing information is provided. The method involves receiving a message; processing the message with a trained artificial neural network based processor, having at least one set of outputs which represent information in a non-arbitrary organization of actions based on an architecture of the artificial neural network based processor and the training; representing as a noise vector at least one data pattern in the message which is incompletely represented in the non-arbitrary organization of actions; and analyzing the noise vector distinctly from the trained artificial neural network.

A single unified machine learning model (e.g., a neural network) is trained to perform both supervised event predictions and unsupervised time-varying clustering for a sequence of events (e.g., a sequence representing a user behavior) using sequences of events for multiple users using a combined loss function. The unified model can then be used for, given a sequence of events as input, predict a next event to occur after the last event in the sequence and generate a clustering result by performing a clustering operation on the sequence of events. As part of predicting the next event, the unified model is trained to predict an event type for the next event and a time of occurrence for the next event. In certain embodiments, the unified model is a neural network comprising a recurrent neural network (RNN) such as an Long Short Term Memory (LSTM) network.

Various embodiments include methods and devices for transforming a data block into weights for a neural network. Some embodiments may include training a first neural network of a cybernetic engram to reproduce the data block, and replacing the data block in memory with weights used by the first neural network to reproduce the data block.

The present design is directed to a system for performing cognitive modeling, including an event acquirer configured to acquire an event comprising an associated date and set of data fields, an analyzer element comprising a plurality of components repeated for each field in an event received from the event acquirer, wherein the analyzer element applies thresholds to each event, determines outliers, evaluates time-ordered behavior, and predicts threshold violations for the event, a periodic set of components configured to operate periodically on demand, the periodic set of components configured to perform peer to peer analysis, actor correlation analysis, actor behavior analysis, semantic rule analysis, and predict rates of change, and a plurality of signal managers interfacing with the analyzer element and the periodic set of components configured to exclude signals based on content properties of data transmitted.

The present design is directed to a system for dispatching cognitive computing across multiple workers, comprising a supervisory node configured to dispatch work to a plurality of compute servers forming interconnected knowledge nodes and a relevancy engine provided in at least one knowledge node. The relevancy engine comprises a rule package and a series of processors organized in a tree arrangement, the series of processors configured to perform functions according to the rule package.

The present design is directed to a series of interconnected compute servers including a supervisory hardware node and a plurality of knowledge hardware nodes, wherein the series of interconnected compute servers are configured to categorize and scale performance of multiple disjoint algorithms across a seemingly infinite actor population, wherein the series of interconnected compute servers are configured to normalize data using a common taxonomy, distribute normalized data relatively evenly across the plurality of knowledge hardware nodes, supervise algorithm execution across knowledge hardware nodes, and collate and present results of analysis of the seemingly infinite actor population.

The present design is directed to a system for defuzzification of multiple qualitative signals into human-centric threat notifications using cognitive computing techniques, including a series of periodic execution components configured to operate over full or partial sets of received data, and a plurality of event reception components configured to operate on each event in a stream relevant to a terrain and selectively provide information to the series of periodic execution components. The system detects and signals anomalies upon detecting behavior inconsistent with normal behavior.

The present design is directed to a cognitive system including a receiver configured to receive a set of actors and associated actor information and receive assets and their associated asset information, a creation apparatus configured to create data dictionary entries for a taxonomy based on the set of actors and the assets and create a cognitive model using the data dictionary entries for a time period, and a computing apparatus configured to compute trust of the cognitive model as a fuzzy number and activate the cognitive model if trust of the cognitive model is above a cognitive model trust threshold. When the cognitive model is activated, the cognitive modeling system is configured to schedule a collection of tasks to run that perform regular extraction of actions from an original data source and perform at least one anomaly analysis associated with the cognitive model.