Abstract: A generative recommender method and system applies trained neural networks to infer related concepts with respect to segments of temporally sequenced content that are inferred to be of particular interest to users. The inferred related concepts of interest may be embodied, for example, in the form vectorized embeddings of natural language and/or images. The embodied inferred related concepts of interest are then input into a generative process that applies trained neural networks to execute one or more vector embedding-based steps that result in generated content elements such as video that are based upon the related concepts of interest.

Abstract: An iterative attention-based neural network training and processing method and system iteratively applies a focus of attention of a trained neural network on syntactical elements and generates probabilities associated with representations of the syntactical elements, which in turn inform a subsequent focus of attention of the neural network, resulting in updated probabilities. The updated probabilities are then applied to generate syntactical elements for delivery to a user. The user may respond to the delivered syntactical elements, providing additional training information to the trained neural network.

Abstract: A temporally sequenced content recommender method and system generates a first vector of affinities from usage information comprising durations of users' attention toward temporally sequenced content. A second vector of affinities is generated by applying a computer-implemented neural network to content objects. The similarity of the vectors of affinities is determined, which informs the generation of recommendations that are delivered to a user. Beneficial serendipity may be incorporated within the generation of the recommendations by, for example, evaluating contrasting affinities in user affinity vectors, by evaluating levels of available behavioral information for users, or by applying randomized or probabilistic methods. Explanations for the recommendations may be provided to users that include inferred user preferences.

Abstract: An integrated neural network and semantic system applies a neural network to interpret an image, determines a syntactical element corresponding to the image in accordance with the interpretation, and determines a first probability that represents a confidence level that the correspondence is accurate. A semantic chain and associated second probability are then generated based on the syntactical element and the first probability, whereby the second probability represents the system's confidence level that the semantic chain accurately reflects objective reality. A natural language communication is generated for delivery to a user that comprises syntactical elements that are in accordance with the semantic chain and the second probability. The communication may further be expected to result in receiving information that will influence the confidence level that the semantic chain accurately reflects objective reality.

Abstract: An optimizing data-to-learning-to-action method and system identifies uncertainties embodied as probability distributions that influence a sequence of decisions. The uncertainties are mapped to a simulation of a computer-based infrastructure that supports the execution of the decisions. Actions with respect to the infrastructure that are expected to reduce the uncertainties are simulated. The probability distributions are updated accordingly for each simulated action and an associated net value of information for each simulated action is generated. The action with the greatest net value of information is implemented and the simulated infrastructure is updated accordingly. The process may then be re-run based upon the updated simulated infrastructure.

Abstract: An auto-learning semantic method and system interprets content by applying neural networks and then generates and/or updates generalized semantic chains based upon the interpretations. The generalized semantics chains have associated weightings or probabilities and the semantic chains can comprise generalizations related to categorization and causation. Automatic learning occurs as the system assess the probabilities associated with the semantic chains and focuses its attention accordingly with the intent of increasing its confidence of its generalizations. The auto-learned generalizations are applied in generating communications that may be directed internally to the system as well as externally.

Abstract: An explanatory integrity determination method and system determines the explanatory integrity of content by analyzing factors that include intentional deception, conscious and unconscious biases, and explanatory gaps. The analyzing is performed by an ensemble of machine learning-based models, including linguistic analysis, semantic chaining, and deep learning. The determined explanatory integrity of an item of content is delivered to a consumer of the content through user interfaces such as a graphical presentations and/or natural language interfaces and/or is applied as an element of decision making by a computer-implemented recommender system.

Abstract: An auto-learning recommender method and system delivers recommendations to users and analyzes the resulting usage behaviors by applying a computer-implemented neural network. Probabilities are automatically determined based upon the analysis that may correspond to inferred preferences. The probabilities inform the generation of additional recommendations that are delivered to users. The generation of the additional recommendations may be further informed by value of information calculations and/or analysis of intrinsic patterns within content.

Abstract: A method and device for temporally sequenced recommendations of activities delivers to users temporally sequenced objects comprising user activities, wherein the delivered objects are selected based, at least in part, on inferences of preferences from usage behaviors. The delivered objects may include activities associated with processor-based devices in addition to human activities. Variations of the system and method include delivering the temporally sequenced objects in accordance with the contents of the objects and user feedback with regard to the objects. Information as to why objects were delivered to users may be provided to the users.

Abstract: A neural network-based content inferencing method and system automatically performs interpretive inferences of content such as video that is associated with a media instance through the application of computer-implemented neural networks. Recommended objects are generated based, at least in part, on the interpretative inferences and are delivered to users. The recommended objects may be further generated based upon inferences of preferences from usage behaviors. User behaviors associated with users interacting with the recommended objects are accessed and elements of the media instance are selected for delivery to users based on an automatic analysis of the user behaviors.

Abstract: A method and system of inferential-based communications applies two sets of values, whereby the first set of values are determined from inferences of the relevancy of textual elements with respect to text-based content and the second set of values are determined from behavioral-based inferences with respect to topics, so as to select appropriate words to be included in communications that are generated for delivery to a user. The inferences that are with respect to text-based content may be determined by applying analytic methods such as Bayesian or statistical learning-based methods. The values that are determined from behavioral-based inferences with respect to topics may correspond to inferences of interest and/or expertise. Words in the communications may be selected and/or arranged in accordance with syntactical rules and may reference elements of the text-based content and/or behavioral inferences.

Abstract: A neural network-based recommender method and system applies computer-implemented neural networks to match information within computer-implemented content objects that contain, for example, text and/or images. The matching of the information is combined with an evaluation of the computer-implemented objects based upon user behaviors to generate computer-implemented topics. Affinities among the topics or among users and the topics are generated based on the matching of the information and the evaluation of the computer-implemented objects. Recommendations are generated for delivery to users based upon the generated topics and affinities. The topics and/or affinities may be based upon inferences of expertise levels.

Abstract: An integrated neural network and semantic system applies a neural network to interpret an image, determines a syntactical element corresponding to the image in accordance with the interpretation, and determines a first probability that represents a confidence level that the correspondence is accurate. A semantic chain and associated second probability are then generated based on the syntactical element and the first probability, whereby the second probability represents the system's confidence level that the semantic chain accurately reflects objective reality. A natural language communication is generated for delivery to a user that comprises syntactical elements that are in accordance with the semantic chain and the second probability. The communication may further be expected to result in receiving information that will influence the confidence level that the semantic chain accurately reflects objective reality.

Abstract: A stream of attention method, system, and apparatus determines a first focus of attention by applying a probabilistic analysis to a first set of syntactical elements. The syntactical elements may be derived from external communications or from communications generated internally by a computer-implemented system. Alternatively, the syntactical elements may be determined based on their association with images accessed from an external or internal source. Actions are performed by the computer-implemented system that are expected to reduce uncertainty with respect to the focus of attention. Such actions may include posing syntactical-based interrogatives directed to external entities or internally, searching a corpus of content, and/or moving to a new location and receiving input at the new location.

Abstract: An inferential media tagging method and system automatically performs inferences from audio content that is associated with a media instance through the application of, for example, Bayesian-based algorithms or computer-implemented neural networks. Recommended objects are generated based, at least in part, on the inferences, and are delivered to users. The recommended objects may be further generated based upon inference tuning controls and inferences of preferences from usage behaviors and may be delivered in a temporal sequence. User behaviors associated with users interacting with the recommended objects are accessed and elements of the media instance are selected for delivery to users based on the user behaviors.

Abstract: A body position-based recommender system infers user preferences from user behaviors and generates adaptive recommendations based upon the inferred user preferences and user body position information. The adaptive recommendations may be delivered kinesthetically. The inferences of user preferences may be based upon mobility inferences. The inferred preferences may be based upon the application of inference weightings that are determined in accordance with usage behavior priority rules that are applied to the user behaviors. Computer-implemented neural networks may be applied to infer user preferences from pictorial-based information. Natural language-based explanations that include the reasoning for the recommendations may be delivered to the user.

Abstract: An inferential-based physical object arrangement method and system infers user preferences from user behaviors and automatically selects computer-implemented objects that represent physical objects based upon the inferred preferences. A media instance is generated for delivery to a user that comprises spatially arranged representations of a selected computer-implemented object and representations of other physical objects that are accessed from a digital map. Computer-implemented neural networks may be applied to infer user preferences by interpreting pictorial-based information and/or to interpret from pictorial-based information the physical objects that are included in the media instances. Natural language-based explanations comprising the reasoning for the delivery of a media instance to a user may be delivered to the user.

Abstract: An optimizing data-to-learning-to-action method and system identifies uncertainties embodied as probability distributions that influence a sequence of decisions. The uncertainties are mapped to a simulation of a computer-based infrastructure that supports the execution of the decisions. Actions with respect to the infrastructure that are expected to reduce the uncertainties are simulated. The probability distributions are updated accordingly for each simulated action and an associated net value of information for each simulated action is generated. The action with the greatest net value of information is implemented and the simulated infrastructure is updated accordingly. The process may then be re-run based upon the updated simulated infrastructure.

Abstract: A peer-to-peer activity sequence structure method and system generates computer-implemented objects comprising a reference to a first system user, a reference to an activity performed by the first system user, and a timestamp associated with the activity performed by the first system user. The computer-implemented objects are syndicated to multiple activity sequence structures on a peer-to-peer basis and then linked to temporally preceding objects within each of the multiple activity sequence structures. An evaluation function that may apply temporal criteria is applied to select a specific activity sequence structure for access and for subsequent linkage to additional computer-implemented objects.

Abstract: A self-assembling learning system and apparatus determines self-assembling actions that are expected to reduce uncertainties that are embodied as probabilities and then updates the probabilities in accordance with information that results from the performing of the self-assembling actions. The updated probabilities inform the determination of subsequent self-assembling actions. Neural networks, simulations of multiple potential self-assembling actions, and expected values of information may be applied in determining the self-assembling actions that are to be performed. Sensors may be applied to receive information that inform the updating of probabilities, and the self-assembling apparatus may be a robotic device. Self-assembling actions may comprise modifications to relationships between elements of a computer-implemented system.