Abstract: According to various embodiments, a machine-learning based system for coronavirus detection is disclosed. The system includes one or more processors configured to interact with a plurality of wearable medical sensors (WMSs). The processors are configured to receive physiological data from the WMSs and questionnaire data from a user interface. The processors are further configured to train at least one neural network based on raw physiological data and questionnaire data augmented with synthetic data and subjected to a grow-and-prune paradigm to generate at least one coronavirus inference model. The processors are also configured to output a coronavirus-based decision by inputting the received physiological data and questionnaire data into the generated coronavirus inference model.

Abstract: A design methodology and tool called INFORM are provided that use a two-phase approach for sample-efficient constrained multi-objective optimization of real-world nonlinear systems. In the first optional phase, one may modify a genetic algorithm (GA) to make the design process sample-efficient, and may inject candidate solutions into the GA population using inverse design methods. The inverse design techniques may be based on (i) a neural network verifier, (ii) a neural network, and (iii) a Gaussian mixture model. The candidate solutions for the next generation are thus a mix of those generated using crossover/mutation and solutions generated using inverse design. At the end of the first phase, one obtains a set of nondominated solutions. In the second phase, one chooses one or more solution(s) from the non-dominated solutions or another reference solution to further improve the objective function values using inverse design methods.

Abstract: According to various embodiments, a machine-learning based system for mental health disorder identification and monitoring is disclosed. The system includes one or more processors configured to interact with a plurality of wearable medical sensors (WMSs). The processors are configured to receive physiological data from the WMSs. The processors are further configured to train at least one neural network based on raw physiological data augmented with synthetic data and subjected to a grow-and-prune paradigm to generate at least one mental health disorder inference model. The processors are also configured to output a mental health disorder-based decision by inputting the received physiological data into the generated mental health disorder inference model.

Abstract: Methods for co-designing transformer-accelerator pairs are provided. The methods may include using a transformer embedding to generate a computational graph and a transformer model. The methods may include running the computational graph through a surrogate model and outputting accuracy data of the surrogate model. The methods may include using an accelerator embedding and the transformer model to simulate training and inference tasks and outputting hardware performance data of the transformer model. The methods may include sending the hardware performance data (such as latency, energy leakage, dynamic energy, and chip area, which may be optimizable performance parameters) and model accuracy data to a co-design optimizer. The methods may include generating an output transformer-accelerator or a transformer-edge-device pair from the co-design optimizer. The transformer model and accelerator embedding may be the output transformer-accelerator or a transformer-edge-device pair.

Abstract: Disclosed is a framework called SCouT that employs a Transformer architecture to make counterfactual predictions that can be used in healthcare and other longitudinal decision-making scenarios. The disclosed approach can use longitudinal donors under an intervention to estimate the synthetic counterfactual for other units. The Transformer-based encoder-decoder model uses a causal map, which enables spatial bidirectionality, to autoregressively generate a synthetic control of a target unit.

Abstract: Systems and methods for tackling a significant problem in data analytics: inaccurate dataset labeling. Such inaccuracies can compromise machine learning model performance. To counter this, label error detection algorithm is provided that efficiently identifies and removes samples with corrupted labels. The provided framework (CTRL) detects label errors in two steps based on the observation that models learn clean and noisy labels in different ways. First, one trains a neural network using the noisy training dataset and obtains the loss curve for each sample. Then, one applies clustering algorithms to the training losses to group samples into two categories: cleanly-labeled and noisily-labeled. After label error detection, one removes samples with noisy labels and retrains the model.

Abstract: According to various embodiments, a method for detecting security vulnerabilities in at least one of cyber-physical systems (CPSs) and Internet of Things (IoT) devices is disclosed. The method includes constructing an attack directed acyclic graph (DAG) from a plurality of regular expressions, where each regular expression corresponds to control-data flow for a known CPS/IoT attack. The method further includes performing a linear search on the attack DAG to determine unexploited CPS/IoT attack vectors, where a path in the attack DAG that does not represent a known CPS/IoT attack vector represents an unexploited CPS/IoT attack vector. The method also includes applying a trained machine learning module to the attack DAG to predict new CPS/IoT vulnerability exploits. The method further includes constructing a defense DAG configured to protect against the known CPS/IoT attacks, the unexploited CPS/IoT attacks, and the new CPS/IoT vulnerability exploits.

Abstract: According to various embodiments, a method for detecting security vulnerabilities in a fifth generation core network (5GCN) is disclosed. The method includes constructing an attack graph from a plurality of regular expressions. Each regular expression corresponds to a sequence of system level operations for a known 5GCN attack. The method further includes performing a linear search on the attack graph to determine unexploited 5GCN attack vectors where path in the attack graph that does not represent a known 5GCN attack vector represents an unexploited 5GCN attack vector. The method also includes applying a trained machine learning module to the attack graph to predict new 5GCN attacks. The trained machine learning module is configured to determine a feasibility of linking unconnected nodes in the attack graph to create a new branch representing a new 5GCN vulnerability exploit.

Abstract: According to various embodiments, a system for detecting security vulnerabilities in at least one of cyber-physical systems (CPSs) and Internet of Things (IoT) devices is disclosed. The system includes one or more processors configured to construct an attack directed acyclic graph (DAG) unique to each CPS or IoT device of the devices. The processors are further configured to generate an aggregate attack DAG from a classification of each device and a location of each device in network topology specified by a system administrator. The processors are also configured to calculate a vulnerability score and exploit risk score for each node in the aggregate attack DAG. The processors are further configured to optimize placement of defenses to reduce an adversary score of the aggregate attack DAG.

Abstract: Devices and methods for processing detected signals at a detector using a processor are provided. The system involves (i) a data compressor that implements an algorithm for converting a set of data into a compressed set of data, (ii) a machine learning (ML) module coupled to the data compressor, the ML module transforming the compressed set of data into a vector and filtering the vector, (iii) a data encryptor coupled to the ML module that encrypts the filtered vector, and (iv) an integrity protection module coupled to the ML module, wherein the integrity protection module protects the integrity of the filtered vector.

Abstract: According to various embodiments, a machine-learning based system for coronavirus detection is disclosed. The system includes one or more processors configured to interact with a plurality of wearable medical sensors (WMSs). The processors are configured to receive physiological data from the WMSs and questionnaire data from a user interface. The processors are further configured to train at least one neural network based on raw physiological data and questionnaire data augmented with synthetic data and subjected to a grow-and-prune paradigm to generate at least one coronavirus inference model. The processors are also configured to output a coronavirus-based decision by inputting the received physiological data and questionnaire data into the generated coronavirus inference model.

Abstract: According to various embodiments, a method for generating one or more optimal neural network architectures is disclosed. The method includes providing an initial seed neural network architecture and utilizing sequential phases to synthesize the neural network until a desired neural network architecture is reached. The phases include a gradient-based growth phase and a magnitude-based pruning phase.

Abstract: According to various embodiments, a machine-learning based system for diabetes analysis is disclosed. The system includes one or more processors configured to interact with a plurality of wearable medical sensors (WMSs). The processors are configured to receive physiological data from the WMSs and demographic data from a user interface. The processors are further configured to train at least one neural network based on a grow-and-prune paradigm to generate at least one diabetes inference model. The neural network grows at least one of connections and neurons based on gradient information and prunes away at least one of connections and neurons based on magnitude information. The processors are also configured to output a diabetes-based decision by inputting the received physiological data and demographic data into the generated diabetes inference model.

Abstract: According to various embodiments, a method for generating a compact and accurate neural network for a dataset that has initial data and is updated with new data is disclosed. The method includes performing a first training on the initial neural network architecture to create a first trained neural network architecture. The method additionally includes performing a second training on the first trained neural network architecture when the dataset is updated with new data to create a second trained neural network architecture. The second training includes growing one or more connections for the new data based on a gradient of each connection, growing one or more connections for the new data and the initial data based on a gradient of each connection, and iteratively pruning one or more connections based on a magnitude of each connection until a desired neural network architecture is achieved.

Abstract: According to various embodiments, a method for detecting security vulnerabilities in at least one of cyber-physical systems (CPSs) and Internet of Things (IoT) devices is disclosed. The method includes constructing an attack directed acyclic graph (DAG) from a plurality of regular expressions, where each regular expression corresponds to control-data flow for a known CPS/IoT attack. The method further includes performing a linear search on the attack DAG to determine unexploited CPS/IoT attack vectors, where a path in the attack DAG that does not represent a known CPS/IoT attack vector represents an unexploited CPS/IoT attack vector. The method also includes applying a trained machine learning module to the attack DAG to predict new CPS/IoT vulnerability exploits. The method further includes constructing a defense DAG configured to protect against the known CPS/IoT attacks, the unexploited CPS/IoT attacks, and the new CPS/IoT vulnerability exploits.

Abstract: According to various embodiments, a method for generating a compact and accurate neural network for a dataset is disclosed. The method includes providing an initial neural network architecture; performing a dataset modification on the dataset, the dataset modification including reducing dimensionality of the dataset; performing a first compression step on the initial neural network architecture that results in a compressed neural network architecture, the first compression step including reducing a number of neurons in one or more layers of the initial neural network architecture based on a feature compression ratio determined by the reduced dimensionality of the dataset; and performing a second compression step on the compressed neural network architecture, the second compression step including one or more of iteratively growing connections, growing neurons, and pruning connections until a desired neural network architecture has been generated.

Abstract: In a system and method for processing detected signals at a detector using a processor, a set of data is converted into a compressed set of data using a compressive sensing component controlled via a processor, the compressed set of data is transformed into a vector and the vector is filtered using a machine learning component controlled via the processor, the filtered vector is encrypted using an encryption component controlled via the processor, and the filtered vector is integrity protected using an integrity protection component controlled via the processor.

Abstract: According to various embodiments, a method for generating one or more optimal neural network architectures is disclosed. The method includes providing an initial seed neural network architecture and utilizing sequential phases to synthesize the neural network until a desired neural network architecture is reached. The phases include a gradient-based growth phase and a magnitude-based pruning phase.

Abstract: According to various embodiments, a method for generating an optimal hidden-layer long short-term memory (H-LSTM) architecture is disclosed. The H-LSTM architecture includes a memory cell and a plurality of deep neural network (DNN) control gates enhanced with hidden layers. The method includes providing an initial seed H-LSTM architecture, training the initial seed H-LSTM architecture by growing one or more connections based on gradient information and iteratively pruning one or more connections based on magnitude information, and terminating the iterative pruning when training cannot achieve a predefined accuracy threshold.

Abstract: According to various embodiments, a stress detection and alleviation (SoDA) system for a user is disclosed. The system includes a SoDA device configured with one or more processors that receive wearable medical sensor (WMS) data from a plurality of WMSs. The processors are programmed to remove one or more artifacts from the WMS data, extract a set of features from the WMS data, remove correlated features from the extracted features to obtain a reduced set of features, classify the reduced set of features in order to determine whether the user is stressed, and generate a response based on whether the user is stressed.