Abstract: A variational hyper recurrent neural network (VHRNN) can be trained by, for each step in sequential training data: determining a prior probability distribution for a latent variable from a prior network of the VHRNN using an initial hidden state; determining a hidden state from a recurrent neural network (RNN) of the VHRNN using an observation state, the latent variable and the initial hidden state; determining an approximate posterior probability distribution for the latent variable from an encoder network of the VHRNN using the observation state and the initial hidden state; determining a generating probability distribution for the observation state from a decoder network of the VHRNN using the latent variable and the initial hidden state; and maximizing a variational lower bound of a marginal log-likelihood of the training data. The trained VHRNN can be used to generate sequential data.

Abstract: Systems 100; devices 110, 120, 130, 150, 160; methods 2400, 2500; and machine-executable programming structures stored in persistent (i.e., non-transitory), computer-readable media 604, 606, 618, 126, 139 for the rapid and secure negotiation, authorization, execution, and confirmation of multi-party data processes, including payment transactions conducted between purchasers 190 having electronic access to bank accounts and other sources of payment, merchants operating e- and/or m-commerce transaction systems 132, 134, 136, and banks and other financial institutions 120 capable of electronically communicating with both.

Abstract: Systems and methods of updating a multi-level data structure for controlling an agent. The method may include: accessing a data structure defining one or more nodes. A non-leaf node of the one or more nodes may be associated with one or more edges for traversing to a subsequent node. An edge of the one or more edges may be associated with a visit count and a softmax state-action value estimation. The method may include identifying a node trajectory including a series of nodes based on an asymptotically converging sampling policy, where the node trajectory includes a root node and a leaf node of the data structure, determining a reward indication associated with the node trajectory; and for at least one non-leaf node, updating the visit count and the softmax state-action value estimate associated with one or more edges of the non-leaf node based on the determined reward indication.

Abstract: A system, non-transitory computer-readable medium, and method are provided. The system comprises at least one processor and memory storing instructions which when executed by the at least one processor configure the at least one processor to perform the method. The non-transitory computer-readable medium has instructions thereon, which when executed by a processor, perform the method. The method comprises determining a similarity score between a first webpage and a second webpage, determining a popularity score of a link between the first webpage and the second webpage, determining a difference between the similarity score and the popularity score, and determining that the link between the first webpage and the second webpage may be improved if the difference is greater than a threshold value.

Abstract: A computer system and method for extending parallelized asynchronous reinforcement learning for training a neural network is described in various embodiments, through coordinated operation of plurality of hardware processors or threads such that each functions as a worker agent that is configured to simultaneously interact with a target computing environment for local gradient computation based on a loss determination and to update global network parameters based at least on local gradient computation to train the neural network through modifications of weighted interconnections between interconnected computing units as gradient computation is conducted across a plurality of iterations of a target computing environment, the loss determination including at least a policy loss term (actor), a value loss term (critic), and an auxiliary control loss. Variations are described further where the neural network is adapted to include terminal state prediction and action guidance.

Abstract: Systems and methods for processing natural language statements. Based on historical records of data associated with an entity, systems and methods provide models for inferring publication of data content associated with the particular entity. The systems and methods may compare newly observed data content to predicted content associated with an entity for evaluating novelty or impact of the newly observed data content.

Abstract: Systems and methods adapted for training a machine learning model to predict data labels are described. The approach includes receiving a first data set comprising first data objects and associated first data labels, and processing, with a user representation model, respective first data objects and associated data labels associated with a unique user representation by fusing the respective first data object and the associated first data labels. First data object representations of the respective first data objects are generated, and the first data object representations and the user representation model outputs are fused to create a user conditional object representation. The machine learning model updates corresponding parameters based on an error value based on a maximum similarity of the projections of the respective user conditional object representation and first data labels in a joint embedding space.

Abstract: A system for generating an adversarial example in respect of a neural network, the adversarial example generated to improve a margin defined as a distance from a data example to a neural network decision boundary. The system includes a data receiver configured to receive one or more data sets including at least one data set representing a benign training example (x); an adversarial generator engine configured to: generate, using the neural network, a first adversarial example (Adv1) having a perturbation length epsilon1 against x; conduct a search in a direction (Adv1-x) using the neural network; and to generate, using the neural network, a second adversarial example (Adv2) having a perturbation length epsilon2 based at least on an output of a search in the direction (Adv1-x).

Abstract: An electronic device and method of correcting bias for supervised machine learning data is provided. The electronic device comprises a processor and memory storing instructions which when executed by the processor configure the processor to perform the method. The method comprises training an auto-encoder with an unbiased subset of historical data, and applying the auto-encoder to correct historical data.

Abstract: Systems, methods, and computer readable media are described to train a compressed neural network with high robustness. The neural network is first adversarially pre-trained with both original data as well as data perturbed by adversarial attacks for some epochs, then “unimportant” weights or filters are pruned through criteria based on their magnitudes or other method (e.g., Taylor approximation of the loss function), and the pruned neural network is retrained with both clean and perturbed data for more epochs.

Abstract: A virtual agent can implement a “chatbot” to provide output based on predictive/prescriptive models for incidents. The virtual agent can integrate with natural language processor for text analysis and summary report generation. The virtual agent can integrate with cognitive search to enable processing of search requests and retrieval of search results. The virtual agent uses computing processes with self-learning systems that use data mining, pattern recognition and natural language processing to mimic the way the human brain works. The virtual agent provides an automated IT system that is capable of resolving incidents without requiring human assistance. The virtual agent can display condensed summaries of a large amount of data and can link the summaries to predictive models and operational risk models to identify risk events and provide summaries of those events.

Abstract: Systems and methods are described in relation to specific technical improvements adapted for machine learning architectures that conduct classification on numerical and/or unstructured data. In an embodiment, two neural networks are utilized in concert to generate output data sets representative of predicted future states of an entity. A second learning architecture is trained to cluster prior entities based on characteristics converted into the form of features and event occurrence such that a boundary function can be established between the clusters to form a decision boundary between decision regions. These outputs are mapped to a space defined by the boundary function, such that the mapping can be used to determine whether a future state event is likely to occur at a particular time in the future.

Abstract: A method for acquiring measurements for a data structure corresponding to an array of variable includes: selecting a subset of elements from the data structure; measuring a sampled value for each of the selected subset of elements; storing each of the sampled values in a K-nearest neighbour (KNN) database and labelling the sampled value as certain; generating a predicted value data structure where each predicted element is generated as the value of its nearest neighbor based on the values stored in the KNN database; for each predicted element: retrieve the predicted element's X nearest neighbours for the sampled value in the KNN database, and when a value of the X nearest neighbours is the same as the predicted element, the predicted element is labelled as certain, otherwise the predicted element is labelled the values as uncertain; and repeating until all elements are labelled as certain.

Abstract: Data drift or dataset shift is detected between training dataset and test dataset by training a scoring function using a pooled dataset, the pooled dataset including a union of the training dataset and the test dataset; obtaining an outlier score for each instance in the training dataset and the test dataset based at least in part on the scoring function; assigning a weight to each outlier score based at least in part on training contamination rates; determining a test statistic based at least in part on the outlier scores and the weights; determining a null distribution of no dataset shift for the test statistic; determining a threshold in the null distribution; and when the test statistic is greater than or equal to the threshold, identifying dataset shift between the training dataset and the test dataset.

Abstract: Systems, devices, methods, and computer readable media are provided in various embodiments having regard to authentication using secure tokens, in accordance with various embodiments. An individual's personal information is encapsulated into transformed digitally signed tokens, which can then be stored in a secure data storage (e.g., a “personal information bank”). The digitally signed tokens can include blended characteristics of the individual (e.g., 2D/3D facial representation, speech patterns) that are combined with digital signatures obtained from cryptographic keys (e.g., private keys) associated with corroborating trusted entities (e.g., a government, a bank) or organizations of which the individual purports to be a member of (e.g., a dog-walking service).

Abstract: Systems and methods are provided to monitor performance of a machine learning model, the method may include steps of: receiving or storing one or more model data sets representative of the machine learning model, wherein the machine learning model has being trained with a first set of training data; analyzing the first set of training data based on one or more performance parameters for the machine learning model, to generate one or more performance data sets; and process the one or more performance data sets to determine one or more values representing a performance of the machine learning model.

Abstract: Systems and methods for responsive stress testing that involve back-end machine learning models that produce enterprise market risk calculations and a front-end interface with graphical elements to visually interact with the machine learning models. The interface can load a particular model, specify ranges of each market variables (shock), and investigate how the market variables impact a portfolio based on the rendered graphical elements.

Abstract: A platform for training deep neural networks using push-to-corner preprocessing and adversarial training. A training engine adds a preprocessing layer before the input data is fed into a deep neural network at the input layer, for pushing the input data further to the corner of its domain.

Abstract: A system receives data associated with a communication between one or more individuals. The data is split between each of the one or more individuals into text associated with that individual. Each of the text is modified to remove stop words and to duplicate key words. The text is merged to form a text corpus, from which a bag of words model is generated. Topics of the bag of words are classified using a topic classifier model. A purpose is identified based on the returned topic and keywords from the topic classifier model. Returned topics and keywords from the topic classifier model are linked to the communication.

Abstract: A smart order router for quantitative trading and order routing and corresponding methods and computer readable media are described. The smart order router includes a machine learning prediction engine configured to, responsive to a control signal received from an upstream trading engine including at least a maximum quantity value and an urgency metric, process input data sets through one or more predictive models to generate the one or more potential combinations of child orders and their associated fill probability metrics, toxicity metrics, and expected gain (loss) metrics and an order placement optimization engine configured to receive the one or more potential combinations of child orders and their associated fill probability metrics, toxicity metrics, and expected gain (loss) metrics and to identify an optimum combination of child orders that maximize an objective function.