A hybrid computer architecture a process providing flexible computing resources across a combination of on-premise computing resources and cloud-based computing resources.
G06F 9/50 - Allocation de ressources, p.ex. de l'unité centrale de traitement [UCT]
G06F 15/16 - Associations de plusieurs calculateurs numériques comportant chacun au moins une unité arithmétique, une unité programme et un registre, p.ex. pour le traitement simultané de plusieurs programmes
G06F 17/00 - TRAITEMENT ÉLECTRIQUE DE DONNÉES NUMÉRIQUES Équipement ou méthodes de traitement de données ou de calcul numérique, spécialement adaptés à des fonctions spécifiques
2.
SYSTEM AND METHOD FOR A MACHINE LEARNING ARCHITECTURE FOR RESOURCE ALLOCATION
A system and method for machine learning architecture for prospective resource allocations are described. The method may include: receiving data records representing historical resource allocations from a user account associated with a first identifier to a resource account associated with a second identifier; deriving input features based on the data records; computing, using a trained neural network architecture, a predicted resource allocation amount and a predicted resource allocation date for the predicted resource allocation amount based on the derived input features; determining, using the trained neural network architecture, a first selection score associated with the predicted resource allocation amount and a second selection score associated with the predicted resource allocation date; and when the first or second selection score is above a minimum threshold, causing to display, at a display device, the associated resource allocation amount or date corresponding to the second identifier.
G06Q 10/04 - Prévision ou optimisation spécialement adaptées à des fins administratives ou de gestion, p. ex. programmation linéaire ou "problème d’optimisation des stocks"
G06F 3/04847 - Techniques d’interaction pour la commande des valeurs des paramètres, p.ex. interaction avec des règles ou des cadrans
G06N 3/0442 - Réseaux récurrents, p.ex. réseaux de Hopfield caractérisés par la présence de mémoire ou de portes, p.ex. mémoire longue à court terme [LSTM] ou unités récurrentes à porte [GRU]
G06Q 10/0631 - Planification, affectation, distribution ou ordonnancement de ressources d’entreprises ou d’organisations
Methods, systems, and techniques for using an actor model payment processing engine to process payments. A payment instruction is received. An event corresponding to the payment instruction is stored in an event journal. The payment processing engine, which is event- sourced and actor- based, perfomis the payment instruction. Perfonning the payment instruction involves transitioning the engine through one or more states in response to the payment instruction, and may involve perfonning actions with non-event sourced and event sourced actors in both stateless and stateful environments.
G06Q 20/40 - Autorisation, p.ex. identification du payeur ou du bénéficiaire, vérification des références du client ou du magasin; Examen et approbation des payeurs, p.ex. contrôle des lignes de crédit ou des listes négatives
G06F 8/656 - Mises à jour pendant le fonctionnement
G06F 11/16 - Détection ou correction d'erreur dans une donnée par redondance dans le matériel
4.
SYSTEMS AND METHODS FOR TOKEN-BASED BROWSER EXTENSION FRAMEWORK
A computer-implemented system and method for orchestrating at least two extensions installed on a browser and for authenticating a user are disclosed. An example method for orchestration includes: receiving, by an extension orchestrator, from a browser launched on a user device, a request from a first extension manager associated with a first extension installed on the browser, the request comprising a first extension ID for the first extension and a second extension ID for a second extension installed on the browser; retrieving, based on the first and second extension IDs, a first extension configuration for the first extension and a second extension configuration for the second extension from a metadata database; and routing a response to the first extension manager, the response comprising the first and second extension configurations and an extension ranking.
H04L 67/63 - Ordonnancement ou organisation du service des demandes d'application, p.ex. demandes de transmission de données d'application en utilisant l'analyse et l'optimisation des ressources réseau requises en acheminant une demande de service en fonction du contenu ou du contexte de la demande
G06F 16/907 - Recherche caractérisée par l’utilisation de métadonnées, p.ex. de métadonnées ne provenant pas du contenu ou de métadonnées générées manuellement
H04L 9/32 - Dispositions pour les communications secrètes ou protégées; Protocoles réseaux de sécurité comprenant des moyens pour vérifier l'identité ou l'autorisation d'un utilisateur du système
A method for deploying a containerized application from a central application management hub to a plurality of cloud environments, the method comprising the steps of: receiving a containerized application suitable for deployment; receiving an environment file designating a first environment and a second environment of the plurality of cloud environments; consulting a routing table to determine a first network path associated with the first environment and a second network path associated with the second environment; generating packets of the containerized application; and sending the packets on the first network path and the second network path; wherein the containerized application is received by a respective operators of the first environment and the second environment for subsequent deployment.
A method for training a neural network utilizing Long Short-Term Memory (LSTM) to model a computer application log as a natural language sequence comprises feeding a training set of application log files to a log file parser, generating, by the log file parser, a set of X application log clusters, where X is a whole number, feeding the whole number X to an untrained LSTM neural network as a hyperparameter representing a number of classes, and training the untrained LSTM neural network using the training set of log files and the hyperparameter X to obtain a trained LSTM neural network.
G06N 3/0442 - Réseaux récurrents, p.ex. réseaux de Hopfield caractérisés par la présence de mémoire ou de portes, p.ex. mémoire longue à court terme [LSTM] ou unités récurrentes à porte [GRU]
Methods, systems, and techniques for agricultural greenhouse gas estimation. Farm data in the form of at least one of revenue generated by a farm, crop information for one or more crops grown on the farm, and land use/farm practice data for land used on the farm to grow the one or more crops is obtained. An emissions estimate is determined based on the obtained data and caused to be displayed to the user via a graphical user interface. A user may be a person responsible for managing multiple farms. That user may be presented with aggregate emissions- related information for all farms, including projected future emissions under various scenarios, and may also iteratively experiment with different farm data values in order to attempt to reduce projected emissions or increase data quality/emissions estimate accuracy.
Methods, systems, and techniques for facilitating proactive recruitment are disclosed, comprising: receiving a user annotation of a candidate profile stored in a database, the user annotation provided by a user; based on at least the received user annotation, determining a sentiment of the user with respect to a candidate associated with the candidate profile; and when the sentiment of the user is determined to be positive, scheduling a notification to be sent to the user in response to a trigger event.
The present disclosure describes an artificial intelligence approach to digital content recommendation where the recommendation mechanics differ based on the amount of information available. In one aspect, a user is identified as an above-threshold user who has consumed at least a threshold number of digital artifacts or a below-threshold user who has consumed fewer digital artifacts and different recommendation engines are used for above-threshold users and below- threshold users. In another aspect, users are bifurcated into low-data users and high-data users. For high-data users, digital artifacts are directly selected, and for low-data users, digital artifacts are indirectly selected by first selecting a digital artifact property criteria and then selecting digital artifacts that satisfy the selected digital artifact property criteria. In another aspect, digital artifacts are selected according to a common recommendation engine, wherein a quantity of digital artifacts consumed by the user is an input to the common recommendation engine.
A procurement system allows a user to provide a request for goods or services. The request is processed to determine its complexity and, for high complexity cases, select an appropriate procurement professional using a trained classifier to handle the procurement request.
Methods, systems, and techniques for data mapping. Company identifiers and an electronic commerce transaction history, such as an online banking transaction history, of a user are retrieved from one or more data repositories. The electronic commerce transaction history includes purchases made from one or more companies identified by the company identifiers. Data mapping is then performed to associate the company identifiers with the purchases represented in the electronic commerce transaction history to identify the companies represented by the company identifiers from which the user made purchases. The company identifiers are then caused to be displayed on a graphical user interface as suggestions to the user as investment suggestions.
A method on applying user data for providing services to a user from a platform of services, the method comprising the steps of: obtaining user profile data pertaining to the user of a network system of an institution; comparing the user profile data to a plurality of different potential life stages in order to determine a selected life stage; identifying one or more services from the platform of services based on the selected life stage; identifying the one or more services to the user via a user interface of a user device; receiving a request from the user through the user device for access to the one or more services; and updating contents of the user profile to include additional profile content related to activity of the user with the one or more services.
G06F 17/00 - TRAITEMENT ÉLECTRIQUE DE DONNÉES NUMÉRIQUES Équipement ou méthodes de traitement de données ou de calcul numérique, spécialement adaptés à des fonctions spécifiques
Computer systems, apparatuses, processors, and non-transitory computer- readable storage devices configured for executing a method for generating proactive advisor recommendation using artificial intelligence. The method has the steps of: partitioning a plurality of clients using a clustering model based on data of the plurality of clients for clustering the plurality of clients into a plurality of client clusters; classifying the clients of at least a first client cluster of the plurality of client clusters into a plurality of client classifications by using one or more random-forest classifiers; and generating financial recommendations for the clients of at least a first client classification of the plurality of client classifications.
Methods, systems, and techniques for facilitating client authentication are disclosed, comprising: receiving an identifier of a client; retrieving client information based on the identifier of the client; assessing a plurality of risk indicators for the client from the client information; determining a risk level for the client based on the plurality of risk indicators; and outputting the risk level for display on a user device.
A data source is monitored. During the monitoring, an arrival at the data source of each of one or more sets of one or more features is detected. In response to detecting the arrival at the data source of at least a first set of one or more features of the one or more sets of one or more features, data is extracted from the first set of one or more features, data for at least a second set of one or more features of the one or more sets of one or more features is estimated, wherein the second set of one or more features has not yet arrived at the data source, and, based on the extracted data and the estimated data, a data quality metric is predicted.
A method for monitoring a network service based on a correlation including network traffic metrics experienced by the network service and infrastructure operational metrics of the network service, the method comprising the steps of: obtaining periodic data including the network traffic metrics, the infrastructure operational metrics, and social media metrics, the social media metrics including content associated with one or more services provided by the network service; storing the network traffic metrics, the infrastructure operational metrics, and social media metrics in a storage for use as historical data representing a predefined period of time; providing a correlation defining a relationship between metrics content of the periodic data; receiving the periodic data during operation of the network service and using the correlation to process the received periodic data to determine an output representing an infrastructure operational metric; comparing the infrastructure operational metric to a predefined operational constraint; generating an alert notification when the infrastructure operational metric contradicts the predefined operational constraint; and sending at least one of the infrastructure operational metric and the alert notification to a support system for subsequent processing.
H04L 43/08 - Surveillance ou test en fonction de métriques spécifiques, p.ex. la qualité du service [QoS], la consommation d’énergie ou les paramètres environnementaux
H04L 43/04 - Traitement des données de surveillance capturées, p.ex. pour la génération de fichiers journaux
A method of generating data on cryptocurrencies is described. Using one or more computer processors, a request to display a benchmark index relating to the cryptocurrencies is received. In response to receiving the request, for each of the cryptocurrencies, a market capitalization value and a price of the cryptocurrency over time are determined. Based on the market capitalization values and the prices over time, the benchmark index is generated and then displayed. In addition, based on the total value of one or more cryptocurrencies over a past period of time, the future price of the one or more cryptocurrencies over the future period of time may be predicted.
A method of making cryptographic key metadata available to key owners while protecting the integrity of the cryptographic key metadata comprises extracting key metadata from a metadata storage on a key data storage system. The metadata storage is logically isolated from a sensitive cryptographic data storage on the key data storage system. The method further comprises transmitting, by unidirectional communication, the extracted key metadata to a user-accessible metadata database that is separate and distinct from the metadata storage on the key data storage system. The method identifies, from the user- accessible metadata database, user-specific metadata for at least one cryptographic key associated with an authorized user associated with the at least one cryptographic key, and communicates the identified user-specific metadata to the authorized user.
H04L 9/32 - Dispositions pour les communications secrètes ou protégées; Protocoles réseaux de sécurité comprenant des moyens pour vérifier l'identité ou l'autorisation d'un utilisateur du système
Methods, systems, and techniques for event notification. An event, such as a payment event that represents a payment transaction having been initiated, completed, or that the transaction is in progress, results in an event engine sending an upstream message to one or more servers. The one or more servers receive the upstream message, which is in a first format. The one or more servers convert the upstream message into a downstream message that is in a second format, such as an ISO 20022 fomiat, and the downstream message is subsequently consumed by an event consumer. The event consumer may consume the downstream message in real-time relative to when the event occurs. Undelivered upstream or downstream messages may be stored in a dead letter channel repository for attempted redelivery.
A method for developing a containerized application using a pipeline platfomi consisting of a plurality of stages with associated development tools, the method comprising the steps of: receiving application parameters and a check-in code for the containerized application; generating a configuration file based on the application parameters, the configuration file containing configuration content including insert code; embedding the insert code into the check-in code; dynamically provisioning an opinionated pipeline based on contents of the configuration file, the opinionated pipeline including the plurality of stages with the associated development tools; setting up one or more control gates in one or more of the plurality of stages; receiving customized code for the containerized application, the customized code representing modifications of the insert code; and packaging the containerized application to include code contents of the check-in code, the customized code, and the insert code; wherein the containerized application is submitted for deployment to one or more environment platfomis upon satisfying the one or more control gates or the containerized application is restricted from the subsequent deployment based on failure of the one or more control gates.
Methods, systems, and techniques for performing automatic source code generation for use in a data transformation process. A computer obtains a data file comprising data transformation rules. Using those rules, the computer automatically generates computer source code for use in a data transformation process to transform source data into target data. The source data may, for example, be raw data from a data lake, and the computer source code may be Scala computer code for execution within an Apache Spark framework. The data lake may execute the computer source code to transform the raw data stored in the data lake into the target data, and the target data may then be stored in the data warehouse.
A method for mapping network connections among a plurality of servers comprises invoking inbuilt OS-native utilities on the servers to identify TCP/IP connections on the servers, parsing the TCP/IP connections into a common representation fomiat, and using the common representation format to map dependencies in the network by differentiating the TCP/IP connections into inbound TCP/IP connections and outbound TCP/IP connections. Local scripts may be used to invoke the inbuilt OS-native utilities and parse the TCP/IP connections into the common representation fonnat.
A method for detecting network anomalies comprises monitoring a network that provides public-facing application services and monitoring at least one external public Internet platfonn outside of the network to obtain volumetric problem report data about the application services. The external public Internet platform is nonspecific to the application services. Responsive to the volumetric problem report data from the external public Internet platform(s) exceeding a threshold, at least one internal network event logging tool is queried for alerts, and from the alerts, at least one anomaly associated with the volumetric problem report data is identified and an anomaly report about the at least one anomaly is generated. Responsive to generating the anomaly report, it may be determined whether the at least one anomaly has a known remediation, and if so, the known remediation may be initiated automatically. Network administrator(s) may also be automatically notified.
A neural network for creating representations of time-series may be trained using a self- supervised approach and as such does not require explicit labelling of the training data. The training uses similarity distillation along both the temporal and instance dimensions. Once trained, the neural network may be used to generate representations of a time- series suitable for use on various downstream tasks.
A process for time-series forecasting is described that decouples stationary conditional distribution modeling from non-stationary dynamic modeling. The forecasting can be applied to non-stationary time-series.
A method for operating a neural network using an encoder-based model to provide a time series forecast, the method comprising: down sampling a time series dataset to generate an initial input having a first scale resolution, such that the first scale resolution is less than a scale resolution of the time series dataset; processing as a first iteration, using the model, the initial input to generate a first output; upsampling by an upsampling function the first output to generate a second input having a second scale resolution, the second scale resolution being higher than the first scale resolution, such that the second input is based on the first output; and processing as a second iteration, using the model, the second input to generate a second output; wherein the second output represents a time series forecast of the time series dataset.
G06Q 10/04 - Prévision ou optimisation spécialement adaptées à des fins administratives ou de gestion, p. ex. programmation linéaire ou "problème d’optimisation des stocks"
G06N 3/047 - Réseaux probabilistes ou stochastiques
27.
SELECTIVE CLASSIFICATION WITH ALTERNATE SELECTION MECHANISM
A method for preparing a trained complete selective classifier can be applied to a trained complete selective classifier having an existing trained selection mechanism. The trained selective classifier is modified to disregard the existing trained selection mechanism and use, as a basis for an alternate selection mechanism, at least one classification prediction value, for example the predictive entropy or the maximum predictive class logit. Optionally, before modifying the trained selective classifier, the method commences with an untrained selective classifier, which may be trained with a modified loss function to obtain the trained selective classifier. The modified loss function has at least one added term, relative to an original loss function, and the at least one added term decreases entropy.
G06V 10/764 - Dispositions pour la reconnaissance ou la compréhension d’images ou de vidéos utilisant la reconnaissance de formes ou l’apprentissage automatique utilisant la classification, p.ex. des objets vidéo
Systems are methods are provided for processing multiple input objectives by a reinforcement learning agent. The method may include: instantiating a reinforcement learning agent that maintains a reinforcement learning neural network and generates, according to outputs of the reinforcement learning neural network, signals for communicating task requests; receiving a plurality of input data representing a plurality of user objectives associated with a task request and a plurality of weights; generating a plurality of preferences based on the plurality of user objectives and the plurality of weights; computing a plurality of loss values; computing a plurality of first gradients based on the plurality of loss values; for a plurality of pairs of references, computing a plurality of similarity metrics; computing an updated gradient based on the first gradients and the plurality of similarity metrics; and updating the reinforcement learning neural network based on the updated gradient.
A computer system and method for populating electronic payment credentials is provided. The system comprises at least one processor and a memory storing instructions which when executed by the processor configure the processor to perform the method. The method comprises receiving a browser extension activation input, sending a payment details request message to a financial institution system, receiving payment details from the financial institution system following authentication at a mobile device, and populating a payment form on the browser using the payment details. Dynamic credentials are provided by the financial institution system and combined with pre-populated tokenized credentials during automatic entry into the payment form.
Systems are methods are provided for processing multiple input objectives by a reinforcement learning agent. The method may include: instantiating a reinforcement learning agent that maintains a reinforcement learning neural network and generates, according to outputs of the reinforcement learning neural network, signals for communicating task requests; receiving a plurality of input data representing a plurality of user objectives associated with a task request; generating, based on the reinforcement learning neural network and the plurality of input data, an action output for generating a signal for communicating the task request; computing a reward based on the action output and the plurality of input data; and updating the reinforcement learning neural network based on the reward.
Systems and methods for generating access entitlements to networked computing resources. Systems may be configured to: receive an input data set representing an entitlement request associated with a user identifier; generate an entitlement prediction associated with the user identifier based on an entitlement model and at least one hierarchical level, the entitlement model defining a cluster representation of entitlement similarity, and wherein the entitlement prediction is based on one or more similarity relationships corresponding to the at least one hierarchical level; and transmit a signal representing the entitlement prediction for granting downstream access to a networked computing resource.
H04L 47/80 - Actions liées au type d'utilisateur ou à la nature du flux
H04L 41/16 - Dispositions pour la maintenance, l’administration ou la gestion des réseaux de commutation de données, p.ex. des réseaux de commutation de paquets en utilisant l'apprentissage automatique ou l'intelligence artificielle
H04L 12/22 - Dispositions pour interdire la prise de données sans autorisation dans un canal de transmission de données
32.
MULTI-MODAL ARTIFICAL NEURAL NETWORK AND A SELF-SUPERVISED LEARNING METHOD FOR TRAINING SAME
A multi-modal artificial neural network and a self-supervised learning method for training that network. The learning method involves processing, using a first modality simple Siamese network, a pair of first modality augmented views of an input; processing, using a second modality simple Siamese network, a pair of second modality augmented views of the input; determining at least one cross-modal loss between the first and second modality simple Siamese networks; determining a total loss from: (i) first and second modality losses respectively determined during the processing using the first and second modality simple Siamese networks; and (ii) the at least one cross-modal loss; and training the first and second modality simple Siamese networks based on the total loss. The trained network may be used to analyze multi-modal content such as video content that has an audio track. A Multi-Modal Multi-Head Network (M3HN) may also be trained to process modality-specific and modality-agnostic representations.
A computing system configured to generate empathy-based machine-learning outputs, which, for example, can include notifications, automatic service delivery, payments, among others. The system receives as inputs a first set of data sets representative of historical behaviour through tracked interactions, a second set of data sets representative of circumstantial knowledge (e.g., environmental factors, such as weather), and a set of empathy model weights from one or more machine learning models that are configured to model one or more empathy consideration components (e.g., curiosity, preconceptions, inspirations, direct experiences, listened experiences, imagination, among others). Corresponding methods and non-transitory computer readable media are contemplated.
A method is provided for training a selective network that includes a selection node for selecting whether to make a prediction. During training, the selection node is reparameterized as a differentiable function of learnable parameters acting on noise from a base distribution. The differentiable function approximates a sampling from a categorical distribution.
An artificial neural network for data imbalanced regression and a method for training that network. A regression dataset is obtained that includes multiple pairs that respectively are made up of inputs and corresponding targets. The inputs are represented in a feature space and the targets are represented in a label space of continuous values. Label space similarities between the targets as represented in the label space are determined, and analogously feature space similarities between the inputs as represented in the feature space are determined. A loss may then be determined based on differences between rankings of the label space similarities and corresponding feature space similarities. That loss may be used to train an artificial neural network.
There is provided a computer system and method for orchestrating user interface, the method include: obtaining a first data set representative of intercepted data communication messages between a user interface of a user and a merchant hosting server; obtaining a second data set representing an instruction set for loading visual elements on the user interface provided from the merchant hosting server; analyzing the first data set to obtain one or more user-specific characteristics; determining if the user-specific characteristics associated with the user satisfy a trigger condition associated with a current resource offering; and responsive to a positive determination: injecting, into the instruction set for loading the visual elements on the user interface provided from the merchant hosting server, code corresponding to an interactive visual element corresponding to the current resource offering.
37.
METHOD AND SYSTEM FOR FACILITATING IDENTIFICATION OF ELECTRONIC DATA EXFILTRATION
Methods, systems, and techniques for facilitating identification of electronic data exfiltration. A message transmission log and screenshot metadata are obtained. A screenshot corresponding to the screenshot metadata is matched to a sent electronic message, such as an email, having a file attachment represented in the message transmission log to generate an event. The screenshot metadata indicates that the screenshot was captured prior to when the message transmission log indicates the electronic message was sent. An anomaly score is determined for the sent electronic message is determined by applying unsupervised machine learning, such as by applying an isolation forest, to score the sent electronic message relative to a baseline. The anomaly score meeting or exceeding an anomaly threshold is treated as potentially being indicative of electronic data exfiltrati on.
38.
METHOD AND SYSTEM FOR DETECTING A CYBERSECURITY BREACH
Methods, systems, and techniques for detecting a cybersecurity breach. The cybersecurity breach may be a synthetic account or an account having been subjected to an account takeover. Electronic account data representative of accounts is obtained in which a first group of the accounts includes accounts flagged as being associated with the breach, and a second group of the accounts includes a remainder of the accounts. The computer system generates from the account data nodes representing the accounts and edges based on account metadata that connect the nodes. The computer system determines, such as by applying a link analysis method to the nodes and edges, a ranking of the accounts of at least part of the second group indicative of a likelihood that those accounts are also associated with the cybersecurity breach. That ranking may be used to identify which of those accounts is also identified with the cybersecurity breach.
39.
SYSTEM AND METHOD FOR SEQUENTIAL DATA PROCESS MODELLING
A system for machine learning architecture for prospective resource allocations. The system may include a processor and a memory. The memory may store processor-executable instructions that, when executed, configure the processor to: receive a sequence of data records representing historical resource allocations from a user associated with a first identifier to another user associated with a second identifier; derive record features based on the sequence of data records representing the historical resource allocations for identifying irregular record features; determine a prospective resource allocation associated with the first identifier and the second identifier based on a neural network model and the derived record features; determine, based on the neural network model, a selection score associated with the prospective resource allocation; and when the selection score is above a minimum threshold, cause to display, at a display device, the prospective resource allocation corresponding to the second identifier.
G06N 3/0442 - Réseaux récurrents, p.ex. réseaux de Hopfield caractérisés par la présence de mémoire ou de portes, p.ex. mémoire longue à court terme [LSTM] ou unités récurrentes à porte [GRU]
G06Q 40/02 - Opérations bancaires, p.ex. calcul d'intérêts ou tenue de compte
40.
SYSTEM AND METHOD FOR DETECTING A BOUNDARY IN IMAGES USING MACHINE LEARNING
A computer-implemented system and method for detecting a boundary in an image are provided. The system includes at least one processor and memory in communication with said at least one processor, wherein the memory stores instructions, when executed at said at least one processor, cause said system to: receive or access a first image comprising a first polygon structure; generate, using a data model representing a neural network, a second image based on the first image by splitting the first polygon structure in the first image, wherein the second image comprises a first portion and a second portion partitioned by a line across the first polygon structure; and generate, based on the second image, a geo-image comprising corresponding spatial-reference information for one or more pixels in the geo-image, the geo-image comprising one of the first portion and the second portion in the second image.
Disclosed are systems, methods, and devices for computing an action for an automated agent. A neural network configured for deep multi-task learning is provided. Each of a subset of layers of the neural network is connected with a respective gating unit configured for dynamically activating or deactivating the respective layer of the neural network. The method includes: receiving, via a communication interface, input data associated with a task type; selecting, from a plurality of layers of a neural network, a subset of layers based on at least the task type; dynamically activating, based on the input data, at least one layer of the subset of layers; and generating an action signal based on a forward pass of the neural network using the dynamically activated at least one layer of the neural network.
A computer-implemented system and method for estimating a Cumulative Distribution Function (CDF) are provided. The method includes: receive input data representing a volume V of a target space indicating a future target event; compute, using the trained neural network, an estimation of a first flux through a boundary of the volume V; compute, using the trained neural network, an estimation of a second flux through a boundary of a volume W of a base space based on the estimation of the first flux through the boundary of the volume V; generate, using the trained neural network, an estimation of a CDF for the volume V based on the second flux through the boundary of the volume W; compute a probability for the future target event based on the estimated CDF for the volume V; and generate a control command based on the probability for the future target event.
A computer-implemented system and method for training a neural network with enforced monotonicity are disclosed. An example system includes at least one processor and memory in communication with said at least one processor, wherein the memory stores instructions for providing a data model representing a neural network for predicting an outcome based on input data, the instructions when executed at said at least one processor causes said system to: receive a feature data as input data; predict an outcome based on the input data using the neural network; compute a loss function based on the predicted outcome and an expected outcome associated with the input data, the loss function f being dependent on a monotonicity penalty fl computed based on a set of training data including the feature data and on a set of random data; and update weights of the neural network based on the loss function.
An automated machine learning approach and toolkit is developed for evaluating the causal impact of an event. This approach includes data generation, optimal model selection, model stability evaluation and model explanation. An example approach includes: generating predictive output data of physical geospatial objects is proposed whereby a first data set representative of geospatial event-based data and a second data set representative of the characteristics of the physical geospatial objects are spatially joined together and utilized to generate a causal graph data model that is then provided for at least one of a trained regression machine learning model, a trained causal machine learning model, and a trained similarity machine learning model to generate the predictive output data representative of event-adjusted characteristics of the physical geospatial objects.
A marketplace for trading bonds on the block chain includes a bond token smart contract that tokenizes the bond for buying/selling using a stablecoin. Each bond generates a corresponding marketplace smart contract. A whitelist smart contract is used to provide permissions for trading bonds on the block chain.
G06Q 40/04 - Transactions; Opérations boursières, p.ex. actions, marchandises, produits dérivés ou change de devises
G06Q 20/06 - Circuits privés de paiement, p.ex. impliquant de la monnaie électronique utilisée uniquement entre les participants à un programme commun de paiement
A method is provided for dynamically visualizing an impact field based on weighted ESG. A portfolio is received, which includes a plurality of assets according to a first configuration, each asset having an associated quantum variable. A raw ESG score is retrieved for each of the assets. A weighted ESG score is detemined for each asset by multiplying the raw ESG score by the quantum variable. A first composite ESG score is fomied by summing the weighted ESG scores for the assets in the first configuration of the portfolio. This is then visually represented by rendering and displaying an impact field having a gradient variable reflective of the first composite ESG score. A recommendation is made for at least one asset in the first configuration. The configuration is changed, another composite ESG score is determined, and the impact field is updated accordingly.
There is described a method of determining whether a fraud claim initiated by a client is legitimate. The method is performed by one or more processors. A fraud claim is received from the client. The fraud claim is in respect of a potentially fraudulent transaction associated with the client. Client data associated with the client is retrieved. The client data includes data relating to historical financial transactions associated with the client. Based on the data relating to the historical financial transactions associated with the client, and based on one or more parameters of the potentially fraudulent transaction, a fraud score associated with the fraud claim is determined. Based on the fraud score, a determination is made as to whether the fraud claim is legitimate.
A method is provided for tracking funds in a real estate transaction using a real estate transaction portal. Through an interface of a real estate transaction portal, a request is accepted from a pre- registered buyer to transfer funds to a pre-registered beneficiary, the funds being in settlement of at least a portion of a real estate transaction. A corresponding payment request is initiated through a digital payment channel. On receipt of a first automated message through the payment channel, the first automated message is decoded as a confirmation of the initiation of the payment request. In real time, a graphical status indicator is displayed to the pre-registered buyer and the pre- registered beneficiary showing the initiation. On receipt of a second automated message through the payment channel, the second automated message is decoded as a completion of the payment request and the graphical status indicator is accordingly updated in real time.
Salient features are extracted from a training data set. The training data set includes, for each of a subset of known legitimate websites and a subset of known phishing websites, Uniform Resource Locators (URLs) and Hypertext Markup Language (HTML) information. The salient features are fed to a machine learning engine, a classifier engine to identify potential phishing websites is generated by applying the machine learning engine to the salient features, and parameters of the classifier engine are tuned. This enables identification of potential phishing websites by parsing a target website into URL information and HTML information, and identifying predetermined URL features and predetermined HTML features. A prediction as to whether the target website is a phishing website or a legitimate website, based on the predetermined URL features and the predetermined HTML features, is received from the classifier engine.
H04L 41/16 - Dispositions pour la maintenance, l’administration ou la gestion des réseaux de commutation de données, p.ex. des réseaux de commutation de paquets en utilisant l'apprentissage automatique ou l'intelligence artificielle
50.
SYSTEM AND METHOD FOR GENERATING AND UPDATING A USER PROFILE FOR AN INSTITUTION BASED ON PEER GROUP DATA
A method for generating a user profile based on a comparison to peer group data, the user being a member of an institution, the method comprising the steps of: obtaining user profile data pertaining to a user of a network service of the institution; accessing group profile data associated with the user; comparing the user profile data to the group profile data to generate comparative data; generating a user profile for presentation on a user interface, the user profile including the comparative data; sending the user profile to the user; receiving a request from the user for a product of institution; and updating the user profile to include information pertaining to the product.
51.
RIGHT-SIZING RESOURCE REQUESTS BY APPLICATIONS IN DYNAMICALLY SCALABLE COMPUTING ENVIRONMENTS
Methods, systems, and techniques for right-sizing resource requests for applications in a dynamically scalable computing environment. In one aspect, a method comprises monitoring resource usage of at least one computer resource by an application executing on a computer system, and monitoring resource requests for the computer resource(s) associated with the application. The method further comprises determining, for the computer resource(s), a resource usage upper bound associated with the application, testing the resource usage upper bound against at least one threshold, determining, from the testing, a resource request adjustment, and dynamically applying the resource request adjustment to the resource requests for the computer resource(s) associated with the application.
An insurance recommendation engine receives customer data and using trained models recommends one or more insurance products that are suitable for the customer. The recommendation engine also provides an explanation as to why the particular products have been recommended. The recommendation models are incorporated into a system that can improves the customer's experience.
Systems, devices, and methods for training an automated agent are disclosed. Multiple automated agents are instantiated, each of the automated agents configured to train over a plurality of training cycles. For each resource, a dedicated portion of a memory device to store state data for the respective resource is allocated. The method includes receiving a request for state data for a particular resource from a subset of the automated agents; for each of the training cycles for the subset of the plurality of automated agents, storing updated state data for the particular resource in the dedicated portion of the memory device allocated to the particular resource; and transmitting an address of the dedicated portion of the memory device for the particular resource to the subset of the automated agents, to facilitate asynchronous reading of the stored state data for the particular resource during each training cycle.
Systems are methods are provided for training an automated agent. The automated agent maintains a reinforcement learning neural network and generates, according to outputs of the reinforcement learning neural network, signals for communicating resource task requests. The system includes a communication interface, a processor, memory, and software code stored in the memory. The software code, when executed, causes the system to: instantiate an automated agent that maintains the reinforcement learning neural network; receive current state data of a resource for a first task; receive historical state metrics of the resource computed based on a plurality of historical tasks; compute normalized state data based on the current state data; and provide the historical state metrics and the normalized state data to the reinforcement learning neural network of said automated agent for training.
55.
SYSTEM AND METHOD FOR MACHINE LEARNING ARCHITECTURE WITH MULTIPLE POLICY HEADS
Systems, devices, and methods for automated generation of resource task requests are disclosed. A reinforcement learning neural network having an output layer with a plurality of policy heads is maintained. At least one reward is provided to the reinforcement learning neural network, the at least one reward corresponding to at least one prior resource task request generated based on outputs of the reinforcement learning neural network. State data are provided to the reinforcement learning neural network, the state data reflective of a current state of an environment in which resource task requests are made. A plurality of outputs is obtained, each from a corresponding policy head, the plurality of outputs including a first output defining a quantity of a resource and a second output defining a cost of the resource. A resource task request signal is generated based on the plurality of outputs from the plurality of policy heads. _ =
Systems, devices, and methods for training an automated agent are disclosed. An automated agent is instantiated. The automated agent includes a reinforcement learning neural network that is trained over a plurality training cycles and provides a policy for generating resource task requests. A learning condition that is expected to impede training of the automated agent during a given training cycle of the plurality of training cycles is detected. In response to the detecting, a disable signal is generated to disable training of the automated agent for at least the given training cycle.
Systems are methods are provided for training an automated agent. The automated agent maintains a reinforcement learning neural network and generates, according to outputs of the reinforcement learning neural network, signals for communicating resource task requests. The system includes a communication interface, a processor, memory, and software code stored in the memory. The software code, when executed, causes the system to: instantiate an automated agent for communicating resource task requests; receive a current feature data structure related to a resource of the resource task requests; maintain a plurality of historical feature data structures related to said resource for a plurality of prior time steps; compute normalized feature data using the current feature data structure and the plurality of historical feature data structures; compute supplemented state data appended with the normalized feature data; and transmit said supplemented state data to the reinforcement learning neural network to train said automated agent.
Systems, methods, and computer readable media are directed in various embodiments for providing multiuser sessions for coordinated electronic transactions. A technical solution is directed to coordinating the electronic transactions across a plurality of instances, where the underlying users of the instances can include at least two users. Access to sensitive information can be restricted using a trusted execution environment and access can be given in accordance with the coordinated electronic transactions.
H04L 65/1094 - Transfert ou partage de sessions entre équipement utilisateurs
G06F 21/57 - Certification ou préservation de plates-formes informatiques fiables, p.ex. démarrages ou arrêts sécurisés, suivis de version, contrôles de logiciel système, mises à jour sécurisées ou évaluation de vulnérabilité
59.
SYSTEM AND METHOD FOR RISK SENSITIVE REINFORCEMENT LEARNING ARCHITECTURE
A computer-implemented system and method for training an auomated agent are disclosed. An example system includes: a communication interface; at least one processor; memory in communication with said at least one processor; software code stored in said memory, which when executed causes said system to: instantiate an automated agent that maintains a reinforcement learning neural network and generates, according to outputs of said reinforcement learning neural network, signals for communicating task requests; receive a plurality of states and a plurality of actions for the automated agent; initialize a learning table Q for the automated agent based on the plurality of states and the plurality of actions; compute a plurality of updated learning tables based on the initialized learning table Q using a utility function, the utility function comprising a monotonically increasing concave function; and generate an averaged learning table Q' based on the plurality of updated learning tables.
A computational approach is proposed herein for controlling a user interface for rendering of interactive graphical control elements representing offers and coupons that are inserted into a computational payment process. In particular, the offers and coupons can interact with stored payment information resident (or tokens thereof) on a digital wallet data structure. The approach can be implemented as a computing system, a computing method operable on a computing system, or a computer program product affixed in the form of a non-transitory computer readable medium storing machine-interpretable instructions.
G06F 3/0484 - Techniques d’interaction fondées sur les interfaces utilisateur graphiques [GUI] pour la commande de fonctions ou d’opérations spécifiques, p.ex. sélection ou transformation d’un objet, d’une image ou d’un élément de texte affiché, détermination d’une valeur de paramètre ou sélection d’une plage de valeurs
G06F 21/62 - Protection de l’accès à des données via une plate-forme, p.ex. par clés ou règles de contrôle de l’accès
G06F 16/22 - Indexation; Structures de données à cet effet; Structures de stockage
A computer system and method for predicting an output for an input are provided. The system comprises at least one processor and a memory storing instructions which when executed by the processor configure the processor to perform the method. The method comprises at least one of estimating a posterior for a plurality of inputs and associated outputs, or providing a point estimate without sampling. The method also comprises predicting the output for a new observation input.
The methods and systems are directed to computational approaches for training and using machine learning algorithms to predict the conditional marginal distributions of the position of agents at flexible evaluation horizons and can enables more efficient path planning. These methods model agent movement by training a deep neural network to predict the position of an agent through time. A neural ordinary differential equation (neural ODE) that represents this neural network can be used to determine the log-likelihood of the agent's position as it moves in time.
A system for machine learning architecture for time series data prediction. The system may be configured to: maintain a data set representing a neural network having a plurality of weights; obtain time series data associated with a data query; generate, using the neural network and based on the time series data, a predicted value based on a sampled realization of the time series data and a normalizing flow model, the normalizing flow model based on a latent continuous-time stochastic process having a stationary marginal distribution and bounded variance; and generate a signal providing an indication of the predicted value associated with the data query.
A system and method for adversarial vulnerability testing of machine learning models is proposed that receives as an input, a representation of a non-differentiable machine learning model, transforms the input model into a smoothed model and conducts an adversarial search against the smoothed model to generate an output data value representative of a potential vulnerability to adversarial examples. Variant embodiments are also proposed, directed to noise injection, hyperparameter control, and exhaustive / sampling-based searches in an effort to balance computational efficiency and accuracy in practical implementation. Flagged vulnerabilities can be used to have models re-validated, re-trained, or removed from use due to an increased cybersecurity risk profile.
G06F 21/57 - Certification ou préservation de plates-formes informatiques fiables, p.ex. démarrages ou arrêts sécurisés, suivis de version, contrôles de logiciel système, mises à jour sécurisées ou évaluation de vulnérabilité
A computer-implemented system and method for training a machine learning model are disclosed, the method includes: maintaining a data set representing a neural network having a plurality of weights; receiving input data comprising a plurality of time series data sets ending with timestamp t-1; generating, using the neural network and based on the input data, a probabilistic forecast distribution prediction at timestamp t and a selection value associated with the probabilistic forecast distribution prediction at timestamp t; computing a loss function based on the selection value; and updating at least one of the plurality of weights of the neural network based on the loss function.
Systems and methods for secure communication of data packets are described using a communications gateway and protocol. One or more payment generator devices utilize trusted execution environments to store identity attestation parameters which are then utilized during registration and/or validation of device identity at the gateway for secure transmission of secure data, including, for example, payment data.
Systems and methods for establishing data linkages are described in various embodiments. A system architecture is described which provides a data processing orchestrator device or service which securely interoperates with data sets at various points in time associated with a set of interactions a user may have with computer systems. The data sets are obtained from different data repositories, and are combined together for analysis such that a first data set representing intents (e.g., web search / browse history) can be combined together with a second data set representing outcomes (e.g., purchase transaction history, web site shopping carts).
A machine learning model is monitored by generating a time series of discrete time bins; for each of the discrete time bins: generating data point labels predicted using a labeling function to apply weak labels to incoming data; for each of the data point labels, generating one or more metric values based on one or more metrics by comparing the data point label to output labels of the machine learning model from the incoming data; and generating an aggregate metric for the time bin based on the one or more metric values for the data point labels of the time bin; and identifying anomalies in the aggregate metrics of the time bins of the time series.
A computer system and method for training a heterogeneous multi-task learning network is provided. The system comprises at least one processor and a memory storing instructions which when executed by the processor configure the processor to perform the method. The method comprises assigning expert models to each task, processing training input for each task, and storing a final set of weights. For each task, weights in the expert models and in gate parameters are initialized, training inputs are provided to the network, a loss is determined following a forward pass over the network, and losses are back propagated and weights are updated for the experts and the gates. At least one task is assigned one exclusive expert model and at least one shared expert model accessible by the plurality of tasks.
Systems and methods for machine learning architecture for out-of-distribution data detection. The system may include a processor and a memory storing processor-executable instructions that may, when executed, configure the processor to: receive an input data set; generate an out-of- distribution prediction based on the input data set and an auto-encoder, the auto-encoder trained based on a pretext task including a transformation of one or more training data sets for reconstruction, the trained auto-encoder trained for reducing a reconstruction error to encode semantic meaning of the training data sets; and generate a signal for providing an indication of whether the input data set is an out-of-distribution data set.
A computer system and method for populating electronic payment credentials is provided. The system comprises at least one processor and a memory storing instructions which when executed by the processor configure the processor to perform the method. The method comprises receiving a browser extension activation input, sending a payment details request message to a financial institution system, receiving payment details from the financial institution system following authentication at a mobile device, and populating a payment form on the browser using the payment details.
A computer-implemented system and method and for learning an entity- independent representation are disclosed. The method may include: receiving an input text; identifying named entities in the input text; replacing the named entities in the input text with entity markers; parsing the input text into a plurality of tokens; generating a plurality of token embeddings based on the plurality of tokens; generating a plurality of positional embeddings based on the respective position of each of the plurality of tokens within the input text; generating a plurality of token type embeddings based on the plurality of tokens and the one or more named entities in the input text; and processing the plurality of token embeddings, the plurality of positional embeddings, and the plurality of token type embeddings using a transformer neural network model to generate a hidden state vector for each of the plurality of tokens in the input text.
DYNAMIC SUBSYSTEM OPERATIONAL SEQUENCING TO CONCURRENTLY CONTROL AND DISTRIBUTE SUPERVISED LEARNING PROCESSOR TRAINING AND PROVIDE PREDICTIVE RESPONSES TO INPUT DATA
A supervised learning processing (SLP) system and method provide cooperative operation of a network of supervised learning processors to concurrently distribute supervised learning processor training, generate predictions, provide prediction driven responses to input objects, and provide operational sequencing to concurrently control and distribute supervised learning processor training and provide predictive responses to input data. The SLP system can dynamically sequence SLP subsystem operations to improve resource utilization, training quality, and/or processing speed. A system monitor-controller can dynamically determine if process environmental data indicates initiation of dynamic subsystem processing sequencing. Concurrently training SLPs provides accurate predictions of input objects and responses thereto and enhances the network by providing high quality value predictions and responses and avoiding potential training and operational delays. The SLP system can enhance the network of SLP subsystems by providing flexibility to incorporate multiple SLP models into the network and train with concurrent commercial operations.
G06F 3/0481 - Techniques d’interaction fondées sur les interfaces utilisateur graphiques [GUI] fondées sur des propriétés spécifiques de l’objet d’interaction affiché ou sur un environnement basé sur les métaphores, p.ex. interaction avec des éléments du bureau telles les fenêtres ou les icônes, ou avec l’aide d’un curseur changeant de comport
G06F 3/04842 - Sélection des objets affichés ou des éléments de texte affichés
75.
SYSTEM AND METHOD FOR DETECTING FRAUDULENT ELECTRONIC TRANSACTIONS
A computer system for, and method of, detecting fraudulent electronic transactions is provided. The system comprises at least one processor and a memory storing instructions which when executed by the processor configure the processor to perform the method. The method comprises accessing a trained model, receiving real-time transaction data, extracting graph- based and statistical features to enrich the real-time transaction data, and determining an account proximity score for the real-time transaction data.
G06Q 20/40 - Autorisation, p.ex. identification du payeur ou du bénéficiaire, vérification des références du client ou du magasin; Examen et approbation des payeurs, p.ex. contrôle des lignes de crédit ou des listes négatives
A computer system and method for answering a natural language question is provided. The system comprises at least one processor and a memory storing instructions which when executed by the processor configure the processor to perform the method. The method comprises receiving a natural language question, generating a SQL query based on the natural language question, generating an explanation regarding a solution to the natural language question as answered by the SQL query, and presenting the solution and the explanation.
Systems and methods for diagnosing and testing fairness of machine learning models based on detecting individual violations of group definitions of fairness, via adversarial attacks that aim to perturb model inputs to generate individual violations. The systems and methods employ auxiliary machine learning models using a local surrogate for identifying group membership and assess fairness by measuring the transferability of attacks from this model. The systems and methods generate fairness indicator values indicative of discrimination risk due to the target predictions generated by the machine learning model, by comparing gradients of the machine learning model to gradients of an auxiliary machine learning model. - 87 -
Systems and methods for adaptively identifying anomalous network communication traffic. The system includes a processor and a memory coupled to the processor. The memory includes processor-executable instructions that configure the processor to: obtain data associated with a sequence of network communication events; determine that the sequence of communication events is generated by a computing agent based on a symmetricity measure associated with the sequence of network communication events; generate a threat prediction value for the sequence of network communication events prior-generated by the computing agent based on a combination of the symmetricity measure and a randomness measure associated with the network communication events; and transmit a signal for communicating that the sequence of network communication events is a potential malicious sequence of network communication events based on the threat prediction value.
A graph structure having nodes and edges is represented as an adjacency matrix, and nodes of the graph structure have node features. A computer-implemented method and system for generating a graph structure are provided, the method comprising: generating an adjacency matrix based on a plurality of node features; generating a plurality of noisy node features based on the plurality of node features; generating a plurality of denoised node features using a neural network based on the plurality of noisy node features and the adjacency matrix; and updating the adjacency matrix based on the plurality of denoised node features.
Systems, methods, and corresponding non-transitory computer readable media describe a proposed system adapted as a platform governing the loading of data in a multiparty secure computing environment. In the multiparty secure computing environment described herein, multiple parties are able to load their secure information into a data warehouse having specific secure processing adaptations that limit both access and interactions with data stored thereon.
G06F 21/62 - Protection de l’accès à des données via une plate-forme, p.ex. par clés ou règles de contrôle de l’accès
G06F 16/90 - Recherche d’informations; Structures de bases de données à cet effet; Structures de systèmes de fichiers à cet effet - Détails des fonctions des bases de données indépendantes des types de données cherchés
81.
SYSTEM AND METHOD FOR MULTIPARTY SECURE COMPUTING PLATFORM
Systems, methods, and corresponding non-transitory computer readable media describe a proposed system adapted as a platform governing the loading of data in a multiparty secure computing environment. In the multiparty secure computing environment described herein, multiple parties are able to load their secure information into a data warehouse having specific secure processing adaptations that limit both access and interactions with data stored thereon.
G06F 21/00 - Dispositions de sécurité pour protéger les calculateurs, leurs composants, les programmes ou les données contre une activité non autorisée
G06F 21/62 - Protection de l’accès à des données via une plate-forme, p.ex. par clés ou règles de contrôle de l’accès
Embodiments generally relate to real-time profile search interfaces and web services with the ability to real-time search for and view details on entities and visualizations of the network. The computer service enables real-time profile search of entities and performs a sequence of data aggregation heuristics to present a consolidated view of an individual.
Systems and methods of dynamic resource allocation. The system may include a processor and a memory coupled to the processor. The memory stores processor-executable instructions that, when executed, configure the processor to: receive a signal representing a resource allocation request; determine a projected resource availability based on a resource model and a second data set including at least one data record unrepresented in batched historical data sets, the batched historical data sets including data records representing at least one of recurring or non- recurring resource allocations, and wherein the resource model is prior- trained based on the batched historical data sets; and generate an output signal for displaying the projected resource availability corresponding with the resource allocation request.
Embodiments relate to web applications and interfaces providing personalized access to relevant wellness resources using microservices and machine learning models. Embodiments relate to web applications and interfaces that provide recommendations based on personas computed using machine learning models. The interfaces and web applications using microservices to provide interface tools that scale to multiple users.
G06F 17/00 - TRAITEMENT ÉLECTRIQUE DE DONNÉES NUMÉRIQUES Équipement ou méthodes de traitement de données ou de calcul numérique, spécialement adaptés à des fonctions spécifiques
Systems and methods for processing extracted data from different data sources to classify the data as an intent, a concern, and an insight for a client using an intent/concern engine. The system has a handler to route the data to a client domain, a financial product domain, a client insight domain and a client concern domain in some embodiments. The system can determine action or task recommendation based on the intent, concern, and insight for the client using a business rule system, and transmits the action or task recommendation to an advisor interface.
A computer system and method for intelligent system diagnostics and management is provided. The system comprises at least one processor and a memory storing instructions which when executed by the processor configure the processor to perform the method. The method comprises receiving resource data and usage data, preprocessing the resource data and the usage data into operational data, training and updating a foresight model using the operational data, receiving a forecast generated by the foresight model, and sending a notification for a recommended action based on the forecast. The forecast may be associated with a future resource state or event associated with the operational data.
Systems and methods of monitoring technology infrastructure using alerts indicative service events and tickets indicative of incidents reported to the support system, including transmitting, to a client via a network, structured support data including issue data and correlation data. The issue data represents issues, which are fewer than the number of tickets, generated by processing textual data of the tickets through a clustering engine implementing a generative probabilistic model and generating the correlation data by associating alerts and tickets by correlating alert-specific identifiers and ticket-specific identifiers. The identifiers are of least one of identifier times, locations, names, or descriptions. A prioritization engine is also disclosed.
Systems and methods for database access monitoring are provided. The system comprises at least one processor and a memory storing instructions which when executed by the at least one processor configure the at least one processor to perform the method. The method comprises receiving login event data, generating a vector representation of a subject entity and a vector representation of an object entity associated with a login event in the login event data, determining a distance between the subject entity and the object entity, and determining an anomaly score for the subject entity and the object entity. The anomaly score based at least in part on the distance between the subject entity and object entity.
G06F 21/50 - Contrôle des usagers, programmes ou dispositifs de préservation de l’intégrité des plates-formes, p.ex. des processeurs, des micrologiciels ou des systèmes d’exploitation
89.
SYSTEM AND METHOD FOR CASCADING DECISION TREES FOR EXPLAINABLE REINFORCEMENT LEARNING
The approaches described herein are adapted to provide a technical, computational mechanism to aid in improving explainability of machine learning architectures or for generating more explainable machine learning architectures. Specifically, the approaches describe a proposed implementation of cascading decision tree (CDT) based representation learning data models which can be structured in various approaches to learn features of varying complexity.
An approach for increasing security of biometric templates is described. An improved system is adapted to split a full set of features or representations of a trained model into a first partial template and a second partial template, the second partial template being stored on a secure enclave accessible only through zero-knowledge proof based interfaces. During verification using the template, a new full set of features is received for comparison, and a model is loaded based on the available portions of the model. Comparison utilizing the second partial template requires the computation of zero-knowledge proofs as direct access to the underlying second partial template is prohibited by the secure enclave.
G06F 21/32 - Authentification de l’utilisateur par données biométriques, p.ex. empreintes digitales, balayages de l’iris ou empreintes vocales
G06Q 20/40 - Autorisation, p.ex. identification du payeur ou du bénéficiaire, vérification des références du client ou du magasin; Examen et approbation des payeurs, p.ex. contrôle des lignes de crédit ou des listes négatives
G07F 19/00 - Systèmes bancaires complets; Dispositions à déclenchement par carte codée adaptées pour délivrer ou recevoir des espèces ou analogues et adresser de telles transactions à des comptes existants, p.ex. guichets automatiques
A system for secure identity data tokenization and processing, the system adapted to receive a loan provisioning request from a merchant computing device associated with an individual and to receive, from a secure identity verification computing device, a secure identity token data object attesting to an identity of the individual. A secure identity token data object is processed to verify the identity of the individual and to initiate a loan origination process. A request for an electronic transaction to be paid or partially paid using funds represented in the dynamic card token data object associated with the unique loan identifier data value and a a payment package data object is generated to encapsulate a tokenized representation of a dynamic card data object associated with the dynamic card token data object, an electronic representation of the funds to be provided from the dynamic card token data object.
G06Q 20/38 - Architectures, schémas ou protocoles de paiement - leurs détails
G06Q 20/40 - Autorisation, p.ex. identification du payeur ou du bénéficiaire, vérification des références du client ou du magasin; Examen et approbation des payeurs, p.ex. contrôle des lignes de crédit ou des listes négatives
92.
SYSTEMS AND METHODS FOR DIVERSE KEYPHRASE GENERATION WITH NEURAL UNLIKELIHOOD TRAINING
Computer implemented methods and systems are provided for generating diverse key phrases while maintaining competitive output quality. A system for training a sequence to sequence (S2S) machine learning model is proposed where neural unlikelihood objective approaches are used at (1) a target token level to discourage the generation of repeating tokens, and (2) a copy token level to avoid copying repetitive tokens from the source text. K-step ahead token prediction approaches are also proposed as an additional mechanism to augment the approach to further enhance the overall diversity of key phrase outputs.
A computer system and method for tokenizing an loT device is provided. The system comprises at least one processor and a memory storing instructions which when executed by the processor configure the processor to perform the method. The method obtaining device information, generating a credential token based on the device information, and processing an electronic transaction using the credential token.
Systems and methods for unsupervised multi-object scene decomposition that involve a spatio- temporal amortized inference model for multi-object video decomposition. Systems and methods involve a new spatio-temporal iterative inference framework to jointly model complex multi-object representations and the explicit temporal dependencies between the frames. Those dependencies improve overall quality of decomposition, encode information about object dynamics and can be used to predict future trajectories of each object separately. Additionally, the model can generate precise estimations and output data even without color information. The model has scene decomposition, segmentation and future prediction capabilities. The processor can use the model to simulate future frames of the scene data.
Systems and methods for neural time series preprocessing and forecasting, dividing time series data to generate chunks of short time series, inputting each of the short time series to a data preprocessing neural network that includes differencing to transform non-stationary data to stationary data and to filter noise, generating and outputting, from the data preprocessing neural network, processed time series data, and inputting the processed time series data to a forecasting neural network. Parameters of the data preprocessing neural network and parameters of the forecasting neural network are learned end-to-end.
ABSTRACT A machine learning failure discriminator machine is described, along with corresponding systems, methods, and non-transitory computer readable media. The approach operates in relation to an iterative machine learning model and includes a phased approach to extract p-values from the iterative machine learning model based on modified versions of the training or validation data sets. The p-values are then used to identify whether various null hypotheses can be rejected, and accordingly, to generate an output data structure indicative of an estimated failure reason, if any. The output data structure may be made available on an API or on a graphical user interface. Date Recue/Date Received 2021-04-09
Systems are methods are provided for facilitating explainability of decision- making by reinforcement learning agents. A reinforcement learning agent is instantiated which generates, via a function approximation representation, learned outputs governing its decision-making. Data records of a plurality of past inputs for the agent are stored, each of the past inputs including values of a plurality of state variables. Data records of a plurality of past learned outputs of the agent are also stored. A group definition data structure defining groups of the state variables are received. For a given past input a given group, data generated reflective of a perturbed input by altering a value of at least one state variable is generated, and are presented to the reinforcement learning agent to obtain a perturbed learned output generated by the reinforcement learning agent; and a distance metric is generated reflective of a magnitude of difference between the perturbed learned output and the past learned output.
Described in various embodiments herein is a technical solution directed to training downstream machine learning models. In particular, specific machines, computer-readable media, computer processes, and methods are described that are utilized to improve data security during training downstream machine learning models, including decreasing the risk of unauthorized access of training data, decreasing the risk of unauthorized use of training data by authorized users, increasing system systemic speed, and reduced overall computational resource requirements. Training data is manipulated prior to being provided for training machine learning models.
A machine learning architecture is proposed that is directed to receive different time-series data sets relating to environmental conditions as well as a target variable for prediction and to transform the time-series data sets for training a plurality of different machine learning models. The trained machine learning models can be utilized to probe various configurations of environmental conditions, and in some embodiments, conduct first and second order co-efficient of variation determinations to generate one or more data values representative of environmental condition sensitivity metrics.
A de-coupled computing infrastructure is described that is adapted to provide domain specific contextual engines based on conversational flow. The computing infrastructure further includes, in some embodiments, a mechanism for directing conversational flow in respect of a backend natural language processing engine. The computing infrastructure is adapted to control or manage conversational flows using a plurality of natural language processing agents.
brevets
