Abstract: Performing an adversarial attack on a neural network classifier is described. A dataset of input-output pairs is constructed, each input element of the input-output pairs randomly chosen from a search space, each output element of the input-output pairs indicating a prediction output of the neural network classifier for the corresponding input element. A Gaussian process is utilized on the dataset of input-output pairs to optimize an acquisition function to find a best perturbation input element from the dataset. The best perturbation input element is upsampled to generate an upsampled best input element. The upsampled best input element is added to an original input to generate a candidate input. The neural network classifier is queried to determine a classifier prediction for the candidate input. A score for the classifier prediction is computed. The candidate input is accepted as a successful adversarial attack responsive to the classifier prediction being incorrect.

Abstract: Performing an adversarial attack on a neural network classifier is described. A dataset of input-output pairs is constructed, each input element of the input-output pairs randomly chosen from a search space, each output element of the input-output pairs indicating a prediction output of the neural network classifier for the corresponding input element. A Gaussian process is utilized on the dataset of input-output pairs to optimize an acquisition function to find a best perturbation input element from the dataset. The best perturbation input element is upsampled to generate an upsampled best input element. The upsampled best input element is added to an original input to generate a candidate input. The neural network classifier is queried to determine a classifier prediction for the candidate input. A score for the classifier prediction is computed. The candidate input is accepted as a successful adversarial attack responsive to the classifier prediction being incorrect.

Abstract: What is disclosed is a system and method for forecasting and imputing an unknown vital measurement of a patient. Temporally successive patient vital measurements are received which comprise irregularly sampled observations {y1, . . . , yN}, where yj denotes the jth observation at time tj, and N is the number of samples. The vital measurements are then provided to a model trained using historical data of patient vital measurements. The model generates a parameter set ?=(A,B,C), where A is a state transition matrix, B is a control matrix, and C is a matrix which maps state-space variables to observation variables. The parameters are used to obtain state-space variable zt which, in turn, is used to forecast an unknown observation yN+1 or to impute an unknown observation yt, where 1<t<N. The historical data is then updated with the forecasted observation yN+1 or the imputed unknown observation yt.

Abstract: Systems and methods of modeling irregularly sampled time series signals with unknown temporal dynamics are disclosed wherein a temporal difference variable (TDV) is introduced to model irregular time differences between subsequent measurements. A hierarchical model is designed comprising two linear dynamical systems that model the effects of evolving TDV on temporal observations. All the parameters of the model, including the temporal dynamics, are statistically estimated using historical data.

Abstract: What is disclosed is a system and method for forecasting and imputing an unknown vital measurement of a patient. Temporally successive patient vital measurements are received which comprise irregularly sampled observations {y1. . . , yN where yN denotes the jth observation at time tj and N is the number of samples. The vital measurements are then provided to a model trained using historical data of patient vital measurements. The model generates a parameter set ?=(A, B, C), where A is a state transition matrix, B is a control matrix, and C is a matrix which maps state-space variables to observation variables. The parameters are used to obtain state-space variable zt which, in turn, is used to forecast an unknown observation yN+1 or to impute an unknown observation yt, where 1<t<N. The historical data is then updated with the forecasted observation yN+1 or the imputed unknown observation yt.

Abstract: Systems and methods of modeling irregularly sampled time series signals with unknown temporal dynamics are disclosed wherein a temporal difference variable (TDV) is introduced to model irregular time differences between subsequent measurements. A hierarchical model is designed comprising two linear dynamical systems that model the effects of evolving TDV on temporal observations. All the parameters of the model, including the temporal dynamics are statistically estimated using historical data.