Abstract: A data source configured to provide a representation of an environment of one or more agents is identified. Using a data set obtained from the data source, a neural network-based reinforcement learning model with one or more attention layers is trained. Importance indicators generated by the attention layers are used to identify actions to be initiated by an agent. A trained version of the model is stored.

Abstract: A machine learning environment utilizing training data generated by customer environments. A reinforced learning machine learning environment receives and processes training data generated by independently hosted, or decoupled, customer environments. The reinforced learning machine learning environment corresponds to machine learning clusters that receive and process training data sets provided by the decoupled customer environments. The customer environments include an agent process that collects training data and forwards the training data to the machine learning clusters without exposing the customer environment. The machine learning clusters can be configured in a manner to automatically process the training data without requiring additional user inputs or controls to configured the application of the reinforced learning machine learning processes.

Abstract: A data source configured to provide a representation of an environment of one or more agents is identified. Using a data set obtained from the data source, a neural network-based reinforcement learning model with one or more attention layers is trained. Importance indicators generated by the attention layers are used to identify actions to be initiated by an agent. A trained version of the model is stored.

Abstract: A simulation management service receives a request to perform reinforcement learning for a robotic device. The request can include computer-executable code defining a reinforcement function for training a reinforcement learning model for the robotic device. In response to the request, the simulation management service generates a simulation environment and injects the computer-executable code into a simulation application for the robotic device. Using the simulation application and the computer-executable code, the simulation management service performs the reinforcement learning within the simulation environment.

Abstract: A training system may create and train a machine learning model with knowledge transfer. The knowledge transfer may transfer knowledge that is acquired by another machine learning model that has been previously trained to the machine learning model that is under training. The knowledge transfer may include a combination of representation transfer and instance transfer, the two of which may be performed alternatingly. The instance transfer may further include a filter mechanism to selectively identify instances with a satisfactory performance to implement the knowledge transfer.

Abstract: A training system may create and train a machine learning model with knowledge transfer. The knowledge transfer may transfer knowledge that is acquired by another machine learning model that has been previously trained to the machine learning model that is under training. The knowledge transfer may include a combination of representation transfer and instance transfer, the two of which may be performed alternatingly. The instance transfer may further include a filter mechanism to selectively identify instances with a satisfactory performance to implement the knowledge transfer.

Abstract: The example embodiments are directed to a system and method for forecasting load flexibility of a power grid. In one example, the method includes receiving temperature values associated with temperature set points of a plurality of loads that are included on a power grid, forecasting a flexibility of the plurality of loads using a polynomial-time mixed-integer non-linear programming (MINLP) optimization based on the received temperature values for the plurality of loads, and outputting information about the forecasted flexibility for display to a display device. The MINLP optimization performs the forecasting of the load flexibility on a fine-grained basis in comparison to conventional methods and is still fast enough that it can be computed in real-time.

Abstract: Techniques for performing image-augmentation based simulations on are described. An exemplary embodiment of such performances includes for each tuple of timestamped image and movement data, generating a next image using an image generation neural network based on the timestamped image and movement data, the image being input into the image generation neural network as a non-rendered image, and generating a reward using a reward generating neural network based on the timestamped image and movement data.

Abstract: A simulation workflow manager obtains a set of parameters for simulation of a system and training of a reinforcement learning model for optimizing an application of the system. In response to obtaining the set of parameters, the simulation workflow manager configures a first compute node that includes a training application for training the reinforcement learning model. The simulation workflow manager also configures a second compute note with a simulation application to perform the simulation of the system in a simulation environment. Data is generated through execution of the simulation in the second compute node that is provided to the first compute node to cause the training application to use the data to train the reinforcement learning model.

Abstract: A simulation application container executes a simulation of a system in a simulation environment, through which an agent representing the system uses a reinforcement learning model to operate within the simulation environment. The simulation application container obtains data indicating how the agent performed in the simulation environment and transmits this data to a robot application container. The robot application container uses the data to update the reinforcement learning model and provides the updated reinforcement learning model to perform another iteration of the simulation and continue training the reinforcement learning model.

Abstract: A simulation management service receives a request to perform reinforcement learning for a robotic device. The request can include computer-executable code defining a reinforcement function for training a reinforcement learning model for the robotic device. In response to the request, the simulation management service generates a simulation environment and injects the computer-executable code into a simulation application for the robotic device. Using the simulation application and the computer-executable code, the simulation management service performs the reinforcement learning within the simulation environment.

Abstract: A method includes receiving a first patient data and a second patient data for a time period, wherein the first patient data and the second patient data are measured from a patient. Further, the method includes identifying a plurality of segmented trends in the first patient data and the second patient data as one of an uptrend, a downtrend, and neutral. Furthermore, the method includes classifying at least one segmented trend from the plurality of segmented trends as a pattern. Additionally, the method includes triggering an alarm as an early warning of patient distress based on the pattern.

Abstract: A method for determining a respiration rate of a subject, includes receiving a first signal and a second signal, each signal being representative of a physiological parameter of the subject. The method includes removing a cardiac artifact signal from the first signal and the second signal to generate a first processed signal and a second processed signal respectively. The method includes removing a motion artifact signal from the first processed signal and the second processed signal to generate a first periodic signal and the second processed signal respectively. The method further includes removing a residual noise signal from the first periodic signal and the second periodic signal to generate a first noise free signal and the second noise free signal respectively. The method includes generating a combined value from a first value and a second value based on the first noise free signal and the second noise free signal respectively.

Abstract: The example embodiments are directed to a system and method for forecasting load flexibility of a power grid. In one example, the method includes receiving temperature values associated with temperature set points of a plurality of loads that are included on a power grid, forecasting a flexibility of the plurality of loads using a polynomial-time mixed-integer non-linear programming (MINLP) optimization based on the received temperature values for the plurality of loads, and outputting information about the forecasted flexibility for display to a display device. The MINLP optimization performs the forecasting of the load flexibility on a fine-grained basis in comparison to conventional methods and is still fast enough that it can be computed in real-time.

Abstract: In one embodiment, a system may include a gas turbine system. the gas turbine system includes a gas turbine, an after-treatment system that may receive exhaust gases from the gas turbine system, and a controller that may receive inputs and model operational behavior of an industrial plant based on the inputs. The industrial plant may include the gas turbine and the after-treatment system. The controller may also determine one or more operational parameter setpoints for the industrial plant, select the one or more operational parameter setpoints that reduce an output of a cost function, and apply the one or more operational parameter setpoints to control the industrial plant.

Abstract: Embodiments of the disclosure are directed to a system for analysis of respiratory distress in hospitalized patients. The system performs multi-parametric simultaneous analysis of respiration rate (RR) and pulse oximetry (SpO2) data trends in order to gauge patterns of patient instability pertaining to respiratory distress. Three patterns in SpO2 and RR are used along with LOWESS algorithm and Chauvenets criteria for outlier rejection to obtain robust short term and long term trends in RR and SpO2. Pattern analysis detects the presence of any one of three pattern types proposed. Further, a learning paradigm is introduced to find unknown instances of respiratory distress. This algorithm in conjunction with the learning model allows early detection of respiratory distress in hospital ward and ICU patients.

Abstract: A system for monitoring respiration, and a method for determining respiration rate, is disclosed. In one embodiment, the respiration rate is determined from a power spectral density template that is updated, or not, based on whether a power spectral density for a current window of in-band filtered impedance respiration signal is determined to be noisy or not.

Abstract: In one embodiment, a system may include a gas turbine system. the gas turbine system includes a gas turbine, an after-treatment system that may receive exhaust gases from the gas turbine system, and a controller that may receive inputs and model operational behavior of an industrial plant based on the inputs. The industrial plant may include the gas turbine and the after-treatment system. The controller may also determine one or more operational parameter setpoints for the industrial plant, select the one or more operational parameter setpoints that reduce an output of a cost function, and apply the one or more operational parameter setpoints to control the industrial plant.

Abstract: A method of prioritizing arrhythmia alarms based on one patient's perfusion level includes receiving arterial blood pressure, electrocardiogram heart rate, and arterial pulse rate values of the one patient during a same time window. Analyzing the set of blood pressure values to determine if an arrhythmia event is indicated, where if an arrhythmia event is indicated, the method includes calculating a systolic blood pressure (SBP) ratio, comparing the SBP ratio to a first predetermined threshold, and if the SBP ratio is less than or equal to the first predetermined threshold, then activating a non-perfusion alarm. If the SBP ratio is greater than the first predetermined threshold, then calculating a standard deviation of a rate differential between the heart rate and the pulse rate values, and if the standard deviation is greater than a second predetermined threshold, then activating the non-perfusion alarm. A system and non-transitory computer media is also presented.

Abstract: A method for determining a respiration rate of a subject, includes receiving a first signal and a second signal, each signal being representative of a physiological parameter of the subject. The method includes removing a cardiac artifact signal from the first signal and the second signal to generate a first processed signal and a second processed signal respectively. The method includes removing a motion artifact signal from the first processed signal and the second processed signal to generate a first periodic signal and the second processed signal respectively. The method further includes removing a residual noise signal from the first periodic signal and the second periodic signal to generate a first noise free signal and the second noise free signal respectively. The method includes generating a combined value from a first value and a second value based on the first noise free signal and the second noise free signal respectively.