Abstract: Systems and methods for predicting road conditions and traffic volume is provided. The method includes generating a graph of one or more road regions including a plurality of road intersections and a plurality of road segments, wherein the road intersections are represented as nodes and the road segments are represented as edges. The method can also include embedding the nodes from the graph into a node space, translating the edges of the graph into nodes of a line graph, and embedding the nodes of the line graph into the node space. The method can also include aligning the nodes from the line graph with the nodes from the graph, and optimizing the alignment, outputting a set of node and edge representations that predicts the traffic flow for each of the road segments and road intersections based on the optimized alignment of the nodes.

Abstract: Methods and systems for detecting and responding to an intrusion in a computer network include generating an adversarial training data set that includes original samples and adversarial samples, by perturbing one or more of the original samples with an integrated gradient attack to generate the adversarial samples. The original and adversarial samples are encoded to generate respective original and adversarial graph representations, based on node neighborhood aggregation. A graph-based neural network is trained to detect anomalous activity in a computer network, using the adversarial training data set. A security action is performed responsive to the detected anomalous activity.

Abstract: A computer-implemented method for efficient and scalable enclave protection for machine learning (ML) programs includes tailoring at least one ML program to generate at least one tailored ML program for execution within at least one enclave, and executing the at least one tailored ML program within the at least one enclave.

Abstract: Systems and methods for a provenance based threat detection tool that builds a provenance graph including a plurality of paths using a processor device from provenance data obtained from one or more computer systems and/or networks; samples the provenance graph to form a plurality of linear sample paths, and calculates a regularity score for each of the plurality of linear sample paths using a processor device; selects a subset of linear sample paths from the plurality of linear sample paths based on the regularity score, and embeds each of the subset of linear sample paths by converting each of the subset of linear sample paths into a numerical vector using a processor device; detects anomalies in the embedded paths to identify malicious process activities, and terminates a process related to the embedded path having the identified malicious process activities.

Abstract: A method for implementing confidential machine learning with program compartmentalization includes implementing a development stage to design an ML program, including annotating source code of the ML program to generate an ML program annotation, performing program analysis based on the development stage, including compiling the source code of the ML program based on the ML program annotation, inserting binary code based on the program analysis, including inserting run-time code into a confidential part of the ML program and a non-confidential part of the ML program, and generating an ML model by executing the ML program with the inserted binary code to protect the confidentiality of the ML model and the ML program from attack.

Abstract: A computer-implemented method for securing software installation through deep graph learning includes extracting a new software installation graph (SIG) corresponding to a new software installation based on installation data associated with the new software installation, using at least two node embedding models to generate a first vector representation by embedding the nodes of the new SIG and inferring any embeddings for out-of-vocabulary (OOV) words corresponding to unseen pathnames, utilizing a deep graph autoencoder to reconstruct nodes of the new SIG from latent vector representations encoded by the graph LSTM, wherein reconstruction losses resulting from a difference of a second vector representation generated by the deep graph autoencoder and the first vector representation represent anomaly scores for each node, and performing anomaly detection by comparing an overall anomaly score of the anomaly scores to a threshold of normal software installation.

Abstract: A computer-implemented method for implementing protocol-independent anomaly detection within an industrial control system (ICS) includes implementing a detection stage, including performing byte filtering using a byte filtering model based on at least one new network packet associated with the ICS, performing horizontal detection to determine whether a horizontal constraint anomaly exists in the at least one network packet based on the byte filtering and a horizontal model, including analyzing constraints across different bytes of the at least one new network packet, performing message clustering based on the horizontal detection to generate first cluster information, and performing vertical detection to determine whether a vertical anomaly exists based on the first cluster information and a vertical model, including analyzing a temporal pattern of each byte of the at least one new network packet.

Abstract: Systems and methods for implementing a system architecture to support a trusted execution environment (TEE) with computational acceleration are provided. The method includes establishing a first trusted channel between a user application stored on an enclave and a graphics processing unit (GPU) driver loaded on a hypervisor. Establishing the first trusted channel includes leveraging page permissions in an extended page table (EPT) to isolate the first trusted channel between the enclave and the GPU driver in a physical memory of an operating system (OS). The method further includes establishing a second trusted channel between the GPU driver and a GPU device. The method also includes launching a unified TEE that includes the enclave and the hypervisor with execution of application code of the user application.

Abstract: A method for ransomware detection and prevention includes receiving an event stream associated with one or more computer system events, generating user-added-value knowledge data for one or more digital assets by modeling digital asset interactions based on the event stream, including accumulating user-added-values of each of the one or more digital assets, and detecting ransomware behavior based at least in part on the user-added-value knowledge, including analyzing destruction of the user-added values for the one or more digital assets.

Abstract: Methods and systems for vehicle fault detection include collecting operational data from sensors in a vehicle. The sensors are associated with vehicle sub-systems. The operational data is processed with a neural network to generate a fault score, which represents a similarity to fault state training scenarios, and an anomaly score, which represents a dissimilarity to normal state training scenarios. The fault score is determined to be above a fault score threshold and the anomaly score is determined to be above an anomaly score threshold to detect a fault. A corrective action is performed responsive the fault, based on a sub-system associated with the fault.

Abstract: Methods and systems for training a neural network model include processing a set of normal state training data and a set of fault state training data to generate respective normal state inputs and fault state inputs that each include data features and sensor correlation graph information. A neural network model is trained, using the normal state inputs and the fault state inputs, to generate a fault score that provides a similarity of an input to the fault state training data and an anomaly score that provides a dissimilarity of the input to the normal state training data.

Abstract: A computer-implemented method for efficient and scalable enclave protection for machine learning (ML) programs includes tailoring at least one ML program to generate at least one tailored ML program for execution within at least one enclave, and executing the at least one tailored ML program within the at least one enclave.

Abstract: Methods and systems for detecting and responding to an intrusion in a computer network include generating an adversarial training data set that includes original samples and adversarial samples, by perturbing one or more of the original samples with an integrated gradient attack to generate the adversarial samples. The original and adversarial samples are encoded to generate respective original and adversarial graph representations, based on node neighborhood aggregation. A graph-based neural network is trained to detect anomalous activity in a computer network, using the adversarial training data set. A security action is performed responsive to the detected anomalous activity.

Abstract: Systems and methods for a provenance based threat detection tool that builds a provenance graph including a plurality of paths using a processor device from provenance data obtained from one or more computer systems and/or networks; samples the provenance graph to form a plurality of linear sample paths, and calculates a regularity score for each of the plurality of linear sample paths using a processor device; selects a subset of linear sample paths from the plurality of linear sample paths based on the regularity score, and embeds each of the subset of linear sample paths by converting each of the subset of linear sample paths into a numerical vector using a processor device; detects anomalies in the embedded paths to identify malicious process activities, and terminates a process related to the embedded path having the identified malicious process activities.

Abstract: Systems and methods for predicting road conditions and traffic volume is provided. The method includes generating a graph of one or more road regions including a plurality of road intersections and a plurality of road segments, wherein the road intersections are represented as nodes and the road segments are represented as edges. The method can also include embedding the nodes from the graph into a node space, translating the edges of the graph into nodes of a line graph, and embedding the nodes of the line graph into the node space. The method can also include aligning the nodes from the line graph with the nodes from the graph, and optimizing the alignment, outputting a set of node and edge representations that predicts the traffic flow for each of the road segments and road intersections based on the optimized alignment of the nodes.

Abstract: Systems and methods for an automotive security gateway include an in-gateway security system that monitors local host behaviors in vehicle devices to identify anomalous local host behaviors using a blueprint model trained to recognize secure local host behaviors. An out-of-gateway security system monitors network traffic across remote hosts, local devices, hotspot network, and in-car network to identify anomalous behaviors using deep packet inspection to inspect packets of the network. A threat mitigation system issues threat mitigation instructions corresponding to the identified anomalous local host behaviors and the anomalous remote host behaviors to secure the vehicle devices by removing the identified anomalous local host behaviors and the anomalous remote host behaviors. Automotive security gateway services and vehicle electronic control units operate the vehicle devices according to the threat mitigation instructions.

Abstract: A computer-implemented method for securing software installation through deep graph learning includes extracting a new software installation graph (SIG) corresponding to a new software installation based on installation data associated with the new software installation, using at least two node embedding models to generate a first vector representation by embedding the nodes of the new SIG and inferring any embeddings for out-of-vocabulary (OOV) words corresponding to unseen pathnames, utilizing a deep graph autoencoder to reconstruct nodes of the new SIG from latent vector representations encoded by the graph LSTM, wherein reconstruction losses resulting from a difference of a second vector representation generated by the deep graph autoencoder and the first vector representation represent anomaly scores for each node, and performing anomaly detection by comparing an overall anomaly score of the anomaly scores to a threshold of normal software installation.

Abstract: Systems and methods are disclosed for securing an enterprise environment by detecting suspicious software. A global program lineage graph is constructed. Construction of the global program lineage graph includes creating a node for each version of a program having been installed on a set of user machines. Additionally, at least two nodes are linked with a directional edge. For each version of the program, a prevalence number of the set of user machines on which each version of the program had been installed is determined; and the prevalence number is recorded to the metadata associated with the respective node. Anomalous behavior is identified based on structures formed by the at least two nodes and associated directional edge in the global program lineage graph. An alarm is displayed on a graphical user interface for each suspicious software based on the identified anomalous behavior.

Abstract: Systems and methods for implementing a system architecture to support a trusted execution environment (TEE) with computational acceleration are provided. The method includes establishing a first trusted channel between a user application stored on an enclave and a graphics processing unit (GPU) driver loaded on a hypervisor. Establishing the first trusted channel includes leveraging page permissions in an extended page table (EPT) to isolate the first trusted channel between the enclave and the GPU driver in a physical memory of an operating system (OS). The method further includes establishing a second trusted channel between the GPU driver and a GPU device. The method also includes launching a unified TEE that includes the enclave and the hypervisor with execution of application code of the user application.

Abstract: A method for implementing confidential machine learning with program compartmentalization includes implementing a development stage to design an ML program, including annotating source code of the ML program to generate an ML program annotation, performing program analysis based on the development stage, including compiling the source code of the ML program based on the ML program annotation, inserting binary code based on the program analysis, including inserting run-time code into a confidential part of the ML program and a non-confidential part of the ML program, and generating an ML model by executing the ML program with the inserted binary code to protect the confidentiality of the ML model and the ML program from attack.