Abstract: Systems and methods for learning based control in accordance with embodiments of the invention are illustrated. One embodiment includes a method for training an adaptive controller. The method includes steps for receiving a set of training data that includes several training samples, wherein each training sample includes a state and a true uncertain effect value. The method includes steps for computing an uncertain effect value based on the state, computing a set of one or more losses based on the true uncertain effect value and the computed uncertain effect value, and updating the adaptive controller based on the computed set of losses.

Abstract: The disclosure provides an approach for learning latent representations of data using factorized variational autoencoders (FVAEs). The FVAE framework builds a hierarchical Bayesian matrix factorization model on top of a variational autoencoder (VAE) by learning a VAE that has a factorized representation so as to compress the embedding space and enhance generalization and interpretability. In one embodiment, an FVAE application takes as input training data comprising observations of objects, and the FVAE application learns a latent representation of such data. In order to learn the latent representation, the FVAE application is configured to use a probabilistic VAE to jointly learn a latent representation of each of the objects and a corresponding factorization across time and identity.

Abstract: A system, method, and apparatus for identifying optimal or near optimal complex stimulation waveforms for a neurostimulator device or neuromodulation device are disclosed. An example method includes using a dueling bandits algorithm with correlation among stimulation arms to select a batch of stimulation arms for sequential application to a patient during a therapy session. Each of the stimulation arms specifies complex stimulation waveform parameter values. Feedback from applying the stimulation arms to the patient is recorded and used to update feedback reward values corresponding to at least some of the stimulation arms using a stimulation arm correlation index. A second batch of stimulations arms is selected based upon the updated feedback reward values and applied to a patient. The method is iteratively repeated over a number of therapy sessions until an optimal or near optimal batch of stimulation arms (defining complex stimulation waveforms) is determined.

Abstract: Systems and methods for learning based control in accordance with embodiments of the invention are illustrated. One embodiment includes a method for training an adaptive controller. The method includes steps for receiving a set of training data that includes several training samples, wherein each training sample includes a state and a true uncertain effect value. The method includes steps for computing an uncertain effect value based on the state, computing a set of one or more losses based on the true uncertain effect value and the computed uncertain effect value, and updating the adaptive controller based on the computed set of losses.

Abstract: A system, method, and apparatus for identifying optimal or near optimal complex stimulation waveforms for a neurostimulator device or neuromodulation device are disclosed. An example method includes using a dueling bandits algorithm with correlation among stimulation arms to select a batch of stimulation arms for sequential application to a patient during a therapy session. Each of the stimulation arms specifies complex stimulation waveform parameter values. Feedback from applying the stimulation arms to the patient is recorded and used to update feedback reward values corresponding to at least some of the stimulation arms using a stimulation arm correlation index. A second batch of stimulations arms is selected based upon the updated feedback reward values and applied to a patient. The method is iteratively repeated over a number of therapy sessions until an optimal or near optimal batch of stimulation arms (defining complex stimulation waveforms) is determined.

Abstract: The disclosure provides an approach for predicting trajectories for real-time capture of video and object tracking, while adhering to smoothness constraints so that predictions are not excessively jittery. In one embodiment, a temporally consistent search and learn (TC-SEARN) algorithm is applied to train a regressor for camera planning. A automatic broadcasting application first receives video input captured by a human-operated camera and another video input captured by a stationary camera with a wide field of view. The automatic broadcasting application extracts feature vectors and pan-tilt-zoom states from the stationary camera input and human-operated camera input, respectively. The automatic broadcasting application further applies the TC-SEARN algorithm to learn a sequential regressor for predicting camera trajectories, based on the extracted feature vectors and pan-tilt-zoom states.

Abstract: The disclosure provides an approach for learning latent representations of data using factorized variational autoencoders (FVAEs). The FVAE framework builds a hierarchical Bayesian matrix factorization model on top of a variational autoencoder (VAE) by learning a VAE that has a factorized representation so as to compress the embedding space and enhance generalization and interpretability. In one embodiment, an FVAE application takes as input training data comprising observations of objects, and the FVAE application learns a latent representation of such data. In order to learn the latent representation, the FVAE application is configured to use a probabilistic VAE to jointly learn a latent representation of each of the objects and a corresponding factorization across time and identity.

Abstract: Approaches are described for discovering a formation associated with an agent group engaging in an activity over a window of time. A formation analysis system computes first and second results for an objective function based on first and second sets of role assignments for each agent in the agent group at first and second moments in time, respectively. The formation analysis system iterates by: replacing the first set of role assignments with the second set of role assignments, and determining whether completion criteria have been met based at least in part on comparing the first result with the second result. If the completion criteria have not been met, then the formation analysis system replaces the second set of role assignments with a third set of role assignments that associate each agent in the first agent group with a different role assignment in the third set of role assignments at a third moment in time.

Abstract: One embodiment provides a method, comprising: training, using deep imitation learning, a neural network associated with a predetermined ghosting model to predict player movements for at least one player during at least one sequence in a game; receiving, at an information handling device, tracking data associated with a player movement path for at least one player during the at least one sequence; analyzing, using a processor, the tracking data to determine at least one feature associated with the at least one player at a plurality of predetermined time points during the at least one sequence; and determining, using the predetermined ghosting model and the at least one feature, a ghosted movement path for the at least one player beginning from one of the plurality of predetermined time points. Other aspects are described and claimed.

Abstract: The disclosure provides an approach for predicting trajectories for real-time capture of video and object tracking, while adhering to smoothness constraints so that predictions are not excessively jittery. In one embodiment, a temporally consistent search and learn (TC-SEARN) algorithm is applied to train a regressor for camera planning. A automatic broadcasting application first receives video input captured by a human-operated camera and another video input captured by a stationary camera with a wide field of view. The automatic broadcasting application extracts feature vectors and pan-tilt-zoom states from the stationary camera input and human-operated camera input, respectively. The automatic broadcasting application further applies the TC-SEARN algorithm to learn a sequential regressor for predicting camera trajectories, based on the extracted feature vectors and pan-tilt-zoom states.

Abstract: Approaches are described for discovering a formation associated with an agent group engaging in an activity over a window of time. A formation analysis system computes first and second results for an objective function based on first and second sets of role assignments for each agent in the agent group at first and second moments in time, respectively. The formation analysis system iterates by: replacing the first set of role assignments with the second set of role assignments, and determining whether completion criteria have been met based at least in part on comparing the first result with the second result. If the completion criteria have not been met, then the formation analysis system replaces the second set of role assignments with a third set of role assignments that associate each agent in the first agent group with a different role assignment in the third set of role assignments at a third moment in time.