Abstract: A system for removing a concept from a trained neural network for executing a classification task, the system comprising: the trained neural network, wherein the trained neural network comprises a hidden layer; and a classifier applied at a layer of the hidden layer, wherein: the classifier defines a representation vector at the layer of the hidden layer, wherein the representation vector classifies instances of the concept and non-instances of the concept at the layer; the classifier defines a concept activation vector, wherein the concept activation vector is a normal vector to the representation vector and the concept activation vector comprises an adversarial penalty objective to reduce the instances of the concept at the layer; and a loss function of the trained neural network is optimised based on a downstream loss of the classification task and the adversarial penalty objective.

Abstract: Embodiments for systems and methods of sequential event prediction with noise-contrastive estimation for marked temporal point process are disclosed.

Abstract: Various embodiments of a computer-implemented system which learns textual representations while filtering out potentially personally identifying data and retaining semantic meaning within the textual representations are disclosed herein.

Abstract: The present disclosure describes techniques for improving data subsampling for recommendation systems. A user-item graph associated with training data may be constructed. An importance of user-item interactions may be estimated via graph conductance based on the user-item graph. An importance of the training data may be measured via sample hardness using a pre-trained pilot model. A subsampling rate may be generated based on the importance estimated from the user-item graph and the importance measured by the pre-trained pilot model.

Abstract: Systems and methods of privacy protection of a user may include compiling an actual number of links in a web history corresponding to each topic in a plurality of internet topics; compiling a topic probability distribution based on the web history; determining an additional number of links to be added to each topic in the plurality of internet topics; and modifying the topic probability distribution by selecting a set of links corresponding to the additional number of links for each topic in the plurality of internet topics.

Abstract: Various embodiments of a computer-implemented system which learns textual representations while filtering out potentially personally identifying data and retaining semantic meaning within the textual representations are disclosed herein.

Abstract: Embodiments of a system and methods for detecting social media designed to spread malicious information to “viral” proportions are disclosed. Historical cascade event data from preselected social media accounts as well as information from related accounts is applied to one or more causality metrics to generate a set of causality values. Causality values are further refined and analyzed to determine how casual a user is with respect to a cascade as opposed to other similar users.

Abstract: Systems and methods of privacy protection of a user may include compiling an actual number of links in a web history corresponding to each topic in a plurality of internet topics; compiling a topic probability distribution based on the web history; determining an additional number of links to be added to each topic in the plurality of internet topics; and modifying the topic probability distribution by selecting a set of links corresponding to the additional number of links for each topic in the plurality of internet topics.

Abstract: Embodiments of a system and methods for detecting social media designed to spread malicious information to “viral” proportions are disclosed. Historical cascade event data from preselected social media accounts as well as information from related accounts is applied to one or more causality metrics to generate a set of causality values. Causality values are further refined and analyzed to determine how casual a user is with respect to a cascade as opposed to other similar users.

Abstract: Embodiments for systems and methods of sequential event prediction with noise-contrastive estimation for marked temporal point process are disclosed.

Abstract: Systems and methods for predicting order-of-magnitude viral cascades in social networks are disclosed.

Abstract: Systems and methods for predicting order-of-magnitude viral cascades in social networks are disclosed.