Abstract: Described is a system for predicting multi-agent movements. A Radon Cumulative Distribution Transform (Radon-CDT) is applied to pairs of signature-formations representing agent movements. Canonical correlation analysis (CCA) components are identified for the pairs of signature-formations. Then, a relationship between the pairs of signature formations is learned using the CCA components. A counter signature-formation for a new dataset is predicted using the learned relationship and a new signature-formation. Control parameters of a device can be adjusted based on the predicted counter signature-formation.

Abstract: Described is a system for predicting the behavior of an autonomous system. The system identifies a taxonomic state of a motion condition of an autonomous vehicle based on a spatiotemporal location of the autonomous vehicle and elements of a driving scenario. Behavior of the autonomous vehicle is predicted based on the taxonomic state of the motion condition. The autonomous vehicle makes and implements a driving operation decision based on the predicted behavior.

Abstract: Described is a system for implicitly predicting movement of an object. In an aspect, the system includes one or more processors and a memory, the memory being a non-transitory computer-readable medium having executable instructions encoded thereon, such that upon execution of the instructions, the one or more processors perform operations of providing an image of a first trajectory to a predictive autoencoder, and using the predictive autoencoder, generating a predicted tactical response that comprises a second trajectory based on images of previous tactical responses that were used to train the predictive autoencoder, and controlling a device based on the predicted tactical response.

Abstract: Described is a system for prediction of adversary movements. In an aspect, the system includes one or more processors and a memory, the memory being a non-transitory computer-readable medium having executable instructions encoded thereon, such that upon execution of the instructions, the one or more processors perform operations of computing relative positions of multiple objects of interest, generating a feature representation by forming a matrix based on the relative positions, predicting movement of the multiple objects of interest by applying clustering to the feature representation and by performing canonical correlation analysis, and controlling a device based on the predicted movement of the multiple objects of interest.

Abstract: Described is a system for enhancing memories during sleep. The system detects, via at least one sensor, brain activity of a user as the user learns a new episode pattern or recalls an episode pattern. The episode pattern is temporally compressed. A specific sleep stage is detected in the user. Upon detection of the specific sleep stage, the system automatically recreates the temporally compressed episode pattern by applying neural stimulation to the user to ensure the consolidation of the episode pattern in the user.

Abstract: A system and method is taught for data processing where an environment around the self-vehicle is encoded into ego centric and geocentric overlapping coordinate systems. The overlapping coordinate systems are then divided into adaptively sized grid cells according to characteristics of environments and the self-vehicle status. Each grid cell is defined with one of representative event patterns and risk values to the self-vehicle. The autonomous driving system is then operative to provide a real time assessment of the surrounding environment in response to the grid cell data. And temporal sequences of the grid cell data are stored in the episodic memory and recalled from it during driving.

Abstract: A system and method is taught for data processing in an autonomous vehicle control system. Using information is acquired from the vehicle, network interface, and sensors mounted on the vehicle, the system can perceive situations around it with much less complexity in computation without losing crucial details, and then make navigation and control decisions. The system and method are operative to generate situation aware events, store them, and recall to predict situations for autonomous driving.

Abstract: Described is a system for an episodic memory used by an automated platform. The system acquires data from an episodic memory that comprises an event database, an event-sequence graph, and an episode list. Using the event-sequence graph, the system identifies a closest node to a current environment for the automated platform. Based on the closest node and using a hash function or key based on the hash function, the system retrieves from the event database an episode that corresponds to the closest node, the episode including a sequence of events. Behavior of the automated platform in the current environment is guided based on the data from the episodic memory.

Abstract: A brain-machine interface for extracting user action intentions within a continuous asynchronous interactive environment is presented. A subliminal stimulus module generates contextually appropriate decision-related stimuli that are unobtrusive to a user. An altered perceptual experience module modifies a user's sensation of the interactive environment based on decision-related stimuli generated from the subliminal stimulus module. A brain monitoring module assesses the user's brain activity in response to the decision-related stimuli and to determine whether an action within the asynchronous interactive environment is intended by the user. Finally, an action is taken based on explicit user input, the user's brain activity in response to the decision-related stimuli, or a combination thereof.

Abstract: Described is system for data classification using formal concept analysis (FCA). In a training phase, the system generates a FCA classification lattice, having a structure, using a set of training data. The set of training data comprises training presentations and classifications corresponding to the training presentations. In a classification phase, a set of test data having classes that are hierarchical in nature is classified using the structure of the FCA classification lattice.

Abstract: Described is a system for feature selection for formal concept analysis (FCA). A set of data points having features is separated into object classes. For each object class, the data points are convolved with a Gaussian function, resulting in a class distribution curve for each known object class. For each class distribution curve, a binary array is generated having ones on intervals of data values on which the class distribution curve is maximum with respect to all other class distribution curves, and zeroes elsewhere. For each object class, a binary class curve indicating for which interval a performance of the known object class exceeds all other known object classes is generated. The intervals are ranked with respect to a predetermined confidence threshold value. The ranking of the intervals is used to select which features to extract from the set of data points in FCA lattice construction.

Abstract: Described is a system for the therapeutic modification of human behavior and, more specifically, a system for the transcranial control of procedural memory reconsolidation for the purposes of enhanced skill acquisition. During operation the system records a practice pattern. The practice pattern is a recording of a sensed brain activity from a sensor array when a subject is performing a skill. The practice pattern is converted to brain activity voxels and stored as both an uncompressed practice pattern and a compressed practice pattern.

Abstract: Described is a system for rank-ordered and cognitive saliency schema-based object selection. The system receives a set of unnormalized probabilities corresponding to a set of objects competing for attentional selection in a current environment. Each unnormalized probability in the set of unnormalized probabilities is based on a likelihood estimation of encountering the corresponding object in the current environment. The set of objects is ranked based on a set of cognitive saliency values corresponding to the set of objects to generate a rank-ordered list of cognitive saliency values. The rank-ordered list of cognitive saliency values is analyzed to detect a schema of the current environment by which the set of objects is ranked. The schema is learned and stored along with a reward measure of the schema's utility. A maximum saliency object in the set of objects is selected based on the rank-ordered list of cognitive saliency values.

Abstract: Described is system for extracting conceptual knowledge representation from neural system. The system extracts a first set of attributes to define a set of objects in a first domain. A first formal concept lattice is constructed comprising the set of objects and the first set of attributes from the first domain. A second set of attributes is extracted to define the set of objects in a second domain. A second formal concept lattice is constructed comprising the set of objects and the second set of attributes from the second domain. The first formal concept lattice is aligned with the second formal concept lattice to link the first set of attributes with the second set of attributes, wherein a combined lattice is generated. The combined lattice is used to relate the first domain to the second domain.

Abstract: Described is a system for action selection based on a combination of neuromodulatory and prefrontal cortex models. The system inputs group attack probability estimates for multiple groups in a prefrontal cortex (PFC) input area of a model instance. The system encodes a dispersion of the group attack probability estimates in an anterior cingulated cortex (ACC) conflict input area of the model instance, resulting in activation of the ACC conflict input area. The activation is propagated to an action area and a neuromodulatory area of the model instance. An action strategy is selected in the action area. The action strategy is implemented, and a reward and a cost is generated for the implemented action strategy. An assessment of possible action strategies is updated based on the generated reward and cost. Each model instance modulates its subsequent action strategy selection based on the updated assessment of the possible action strategies.

Abstract: Described is a system for optimizing rapid serial visual presentation (RSVP). A similarity metric is computed for RSVP images, and the images are sequenced according to the similarity metrics. The sequenced images are presented to a user, and neural signals are received to detect a P300 signal. A neural score for each image is computed, and the system is optimized to model the neural scores. The images are resequenced according a predictive model to output a sequence prediction which does not cause a false P300 signal. Additionally, the present invention describes computing a set of motion surprise maps from image chips. The image chips are labeled as static or moving and prepared into RSVP datasets. Neural signals are recorded in response to the RSVP datasets, and an EEG score is computed from the neural signals. Each image chip is then classified as containing or not containing an item of interest.

Abstract: Described is a system for adaptive memory recall. The system receives original input data, then divides the original input data into multiple data groups to serve as input for a network comprising a pattern separation layer and an autoassociative memory layer. The original input data is processed with a pattern separation component of the pattern separation layer, and each pattern separation component generates an increased-contrast version of the original input data it processes. The generated increased-contrast version of the original input data is combined with the original input data and stored in an autoassociative memory component of the autoassociative memory layer for each data group. New input data is received, and a parameter that controls processing in the pattern separation layer is modulated to determine an optimal parameter, which indicates that a memory recall between the stored data and the new set of input data is achieved.

Abstract: Described is a system for flexible cognitive perception and selection. A pre-processing recognition module filters and tags input data from an environment resulting in a tagged percept. The tagged percept is stored and associated with a knowledge frame by a memory module based on shared descriptors, resulting in an activated knowledge frame. A utility rating is then supplied to each activated knowledge frame based on a set of reward values by an evaluation module. The activated knowledge frames are sorted, compared, and evaluated for a goodness of fit between the utility ratings of the activated knowledge frames and the input data by a hypothesis module. A best hypothesis for a current situation in the environment is determined based on a current highest rated activated knowledge frame.

Abstract: The present invention relates to a system for identifying regions of interest in visual imagery. The system is configured to receive a series of consecutive frames representing a scene as captured from N sensors. The frames include at least a current frame and a previous frame. A surprise map can be generated based on features found in the current frame and the previous frame. The surprise map having a plurality of values corresponding to spatial locations within the scene. Based on the values, a surprise in the scene can be identified if a value in the surprise map exceeds a predetermined threshold.

Abstract: Described is a system for optimizing rapid serial visual presentation (RSVP). A similarity metric is computed for RSVP images, and the images are sequenced according to the similarity metrics. The sequenced images are presented to a user, and neural signals are received to detect a P300 signal. A neural score for each image is computed, and the system is optimized to model the neural scores. The images are resequenced according a predictive model to output a sequence prediction which does not cause a false P300 signal. Additionally, the present invention describes computing a set of motion surprise maps from image chips. The image chips are labeled as static or moving and prepared into RSVP datasets. Neural signals are recorded in response to the RSVP datasets, and an EEG score is computed from the neural signals. Each image chip is then classified as containing or not containing an item of interest.