Patents.

Machine learning based tenant-specific chatbots for performing actions in a multi-tenant system

A multi-tenant system performs custom configuration of a tenant-specific chatbot to process and act upon natural language requests. The multi-tenant system configures the tenant-specific chatbots without requiring tenant-specific training. The multi-tenant system providing a user interface for configuring a tenant-specific set of permitted actions. The multi-tenant system determines a set of example phrases for each of the selected permitted actions. The multi-tenant system receives a natural language request from a user and identifies the action that the user wants to perform. The multi-tenant system uses a neural network to compare the natural language request with example phrases to identify an example phrase that matches the natural language request. The multi-tenant system performs the action corresponding to the matching example phrase.

Protein and gene biomarkers for rejection of organ transplants

Metadata chain of trust

A system may receive digital content from a first user and verify that the first user is an owner of the digital content. The system may also generate, based upon a request by the first user, a first proxy for a portion of the digital content granting access to the portion of the digital content to a second user, and generate, based upon a request by the second user, a second proxy granting access to a third user.

Leveraging dialogue history in updated dialogue

Attention-based deep reinforcement learning for autonomous agents

A data source configured to provide a representation of an environment of one or more agents is identified. Using a data set obtained from the data source, a neural network-based reinforcement learning model with one or more attention layers is trained.

Cryptoasset custodial system with proof-of-stake blockchain support

Methods and systems for secure storage and retrieval of information, such as private keys, useable to control access to a blockchain, include: receiving, in a cryptoasset custodial system, a request to authorize a staking operation associated with a blockchain, wherein the staking operation is...

Scheduled federated learning for enhanced search

An indication of availability over time and resource usage is maintained for each computing device of a plurality of computing devices. An optimal combination of a subset of the plurality of computing devices is determined for each round of one or more rounds of training based on the availability over time and the resource usage for each computing device. A global model is generated utilizing the one or more optimal combinations of the plurality of computing devices and a query is performed utilizing the global model.

Reducing description length based on confidence

Systems and processes for operating an intelligent automated assistant are provided. In one example, a user request for a media item is received. Based on the user request, at least one media item and a description of the at least one media item are identified. A confidence level is obtained that an identified media item of the at least one media item corresponds to the requested media item. In accordance with a determination that the confidence level exceeds a first confidence threshold, a length of the identified description is reduced to obtain a modified description and the modified description of the identified media item is provided in a first spoken response.

System and methods for training task-oriented dialogue (TOD) language models

Secure computer-implemented execution and evaluation of programming assignments for on demand courses

A computer system includes a course repository and a server. The course repository includes a video of an online course, an instructor provided output for a programming assignment associated with the online course, and a submission script for generating student provided output for the programming assignment. The server includes a submission module that receives and verifies the student provided output for the programming assignment. The server further includes one or more built-in grader applications that compare the student provided output with the instructor provided output to determine a grade for the programming assignment.

Parking spot locator based on personalized predictive analytics

In an approach to creating and using a reinforcement learning model for personalizing a recommendation of a parking spot, one or more computer processors receive a first destination associated with the vehicle. One or more computer processors determine a parking spot availability in proximity to the destination. One or more computer processors determine a recommended parking spot location. One or more computer processors display the recommended parking spot location to the user. One or more computer processors determine a first parking spot selection. One or more computer processors receive a first satisfaction rating associated with the recommended parking spot location.

Load balancing computing resources in an observability pipeline system

Avatar Personalization Using Image Generation

Prediction-correction approach to zero shot learning

Process control tool for processing big and wide data

A process control tool for processing wide data from automated manufacturing operations. The tool including a feature selector, an analysis server, and a visualization engine. The feature selector receives process input data from at least one manufacturing process application, wherein the process input data includes a plurality of observations and associated variables, converts the received process input data to a stacked format having one row for each variable in each observation, converts identified categorical variables into numerical variables and identified time-series data into fixed numbers of intervals, computes statistics that measure the strengths of relationships between predictor values and an outcome variable, orders, filters, and pivots the predictor values. The analysis server performs at least one operation to identify interactions between predictor values, e.g. using maximum Likelihood computations or predefined searches, in the filtered predictor values. The visualization engine displays the interactions for use in managing the manufacturing operations.

Expert Detection in Social Networks

Digital credentials for primary factor authentication

A system for credential authentication includes and interface and a processor. The interface is configured to receive a request for authorization to access from an application. The processor is configured to determine a set of credentials that can enable authorization to access; generate a proof request challenge; receive a proof response; determine that the proof response is valid based at least in part on information stored in a distributed ledger; generate a token; and provide the token.

Virtual dialog system performance assessment and enrichment

Embodiments are provided that relate to a computer system, a computer program product, and a computer-implemented method for improving performance of a virtual dialog agent system employing an automated virtual dialog agent. Embodiments involve generating ground truth (GT) from a user's knowledge base, and leveraging the GT to evaluate how the virtual dialog agent performs with the GT. The evaluation measures quality of a multi-turn virtual dialog, and generates a remediation plan directed at an algorithmic improvement of the virtual dialog agent.

Image augmentation and object detection

Computing systems may support image classification and image detection services, and these services may utilize object detection/image classification machine learning models. The described techniques provide for normalization of confidence scores corresponding to manipulated target images and for non-max suppression within the range of confidence scores for manipulated images. In one example, the techniques provide for generating different scales of a test image, and the system performs normalization of confidence scores corresponding to each scaled image and non-max suppression per scaled image These techniques may be used to provide more accurate image detection (e.g., object detection and/or image classification) and may be used with models that are not trained on modified image sets. The model may be trained on a standard (e.g. non-manipulated) image set but used with manipulated target images and the described techniques to provide accurate object detection.

Extensible and elastic data management services engine external to a storage domain

A data management services architecture includes architectural components that run in both a storage and compute domains. The architectural components redirect storage requests from the storage domain to the compute domain, manage resources allocated from the compute domain, ensure compliance with a policy that governs resource consumption, deploy program code for data management services, dispatch service requests to deployed services, and monitor deployed services. The architectural components also include a service map to locate program code for data management services, and service instance information for monitoring deployed services and dispatching requests to deployed services. Since deployed services can be stateless or stateful, the services architecture also includes state data for the stateful services, with supporting resources that can expand or contract based on policy and/or service demand. The architectural components also include containers for the deployed services.

System for credential storage and verification

A system for credential storing and verifying includes an interface and a processor. The interface is configured to receive an indication to register a credential. The processor is configured to indicate to store in a distributed ledger a DID document associated with a holder identifier using a smart contract. Storing using the smart contract employs a dual signature authentication scheme to authorize storing based at least in part on an individual signature and a ledger writer signature. The processor is further configured to indicate to store in the distributed ledger a schema associated with an issuer of the credential using the smart contract and indicate to store in the distributed ledger a credential definition associated with the schema using the smart contract.

Digital credentials for access to sensitive data

A system for providing access is configured to receive an application access request from an application for authorization to access and a sensitive data access request from the application for authorization to access a document that includes sensitive data. The system is further configured to determine to authorize access to the application in response to the application access request; to determine the user authentication device in response to the sensitive data access request; to provide a secondary request for authorization to access sensitive data to the user authentication device in response to the sensitive data access request, receive a secondary request response from the user authentication device to the secondary request; and to provide the secondary request response to the application enabling access to the sensitive data, where the document is encrypted for delivery to the application for the user using a blinding secret and an identity private key.

Systems and methods for pixel-based quantum state visualization

Sensitive data service storage

Providing flexible service access using identity provider

Cryptoasset custodial system with vault-specific rules governing different actions allowed for different vaults

Methods, and systems for secure storage and retrieval of information, such as private keys, useable to control access to a blockchain, include: receiving a request to take an action with respect to a vault of multiple different vaults in a cryptoasset custodial system, and each of the multiple...

Telematics data processing for collision detection

Semantic representations using structural ontology for assistant systems

Weakly supervised natural language localization networks for video proposal generation

Route safety determination system

Systems and methods are provided for receiving location data for a first location and a second location and generating a plurality of candidate routes to travel from the first location to the second location, based on the location data, each candidate route comprising a plurality of segments. The systems and methods further generate a safety score for each segment of each candidate route of the plurality of candidate routes, generate a safety score for each candidate route based on safety scores generated for each segment associated with each candidate route, select a best candidate route using the safety score associated with each of the candidate routes, and provide a recommendation for a travel route comprising the best candidate route.

Digital credentials for location aware check in

A system for credential authentication include an interface configured to receive a create indication to create a location aware credential, wherein the location aware credential specifies visit location data and receive a check in indication to check in from an authentication device, wherein the authentication device provides the check in indication to check in in response to determining that a detected location is within a geographic boundary designated in the visit location data of the location aware credential, and a processor configured to provide a proof request, receive a proof response, validate the proof response using a distributed ledger, and provide a success indication of successful check in.

Automated mechanisms to resolve explore-exploit dilemma with adaptive revival of dormant content elements

A system including one or more processors and one or more non-transitory computer-readable media storing computing instructions configured to run on the one or more processors and perform: displaying content elements on one or more websites to users; tracking respective impression response data.

Message objection identification and handling

A system uses a data stream connection to detect a new message between users, applies pre-processing to identify it as an objection message candidate, and retrieves responses to handle the objection.

Vehicle sensor system and method of use

A system and method for collecting and processing sensor data for facilitating and/or enabling autonomous, semi-autonomous, and remote operation of a vehicle, including: collecting surroundings at one or more sensors, and determining properties of the vehicle surroundings from the collected sensor data.

Secure distribution of device key sets over a network

A system is provided for distribution of device key sets over a network in a protected software environment (PSE). In the system, a client device includes a connection interface for receiving a crypto hardware (CH) token. Continuation of US11063753B2.

Systems and methods for learning for domain adaptation

Personalized mechanisms to resolve explore-exploit dilemma with dynamically shared learnings

A system including one or more processors and one or more non-transitory computer-readable media storing computing instructions configured to run on the one or more processors and perform: displaying content elements on one or more websites to users; performing a classification of the users.

Robustness evaluation via natural typos

Embedding drift in a machine learning model

Automatic machine learning model generation

A system may automatically generate a predictive machine learning model by automatically performing various processes based on an analysis of the data as well as metadata associated with the data.

Design-time information based on run-time artifacts in transient cloud-based distributed computing clusters

Distinct value estimation for query planning

Design-time information based on run-time artifacts in a distributed computing cluster

Cloud platform access system

A system includes an interface and a processor. The interface is configured to receive, at an application routing platform, an API call for an application platform comprising a signed tenant token. The processor is configured to determine that the signed tenant token is valid; determine an application platform token for the application platform; associate a root certificate with the application platform token; determine routing information to the application platform based at least in part on the API call; and provide the application platform the API call and the application platform token using the routing information to enable access to the application platform, wherein the application platform determines whether the application platform token is valid using the root certificate and executes the API call in response to a determination that the application platform token is valid.

Semantic representations using structural ontology for assistant systems

Fault tolerant autonomous vehicle platform

Technologies are provided for detecting anomalous data at runtime in an autonomous vehicle component that is indicative of a byzantine fault, and providing real-time remediation. Input and output data streams of the vehicle component can be analyzed to generate a normal operational model for the vehicle component. Using the model, deviations from a known steady state of operation of the vehicle component can be detected and flagged as potential faults. The vehicle component can then be isolated and/or restarted. Additionally or alternatively, a specification can be defined that specifies allowed component interactions in the autonomous vehicle system. The specification can be used to generate tests that can be used to validate the functional correctness of the vehicle components.

Automatic machine learning model generation

A system may automatically generate a predictive machine learning model by automatically performing various processes based on an analysis of the data as well as metadata associated with the data. The system may accept a selection of data and a prediction field from the data.

Parameter sharing in federated learning

One embodiment provides a method for federated learning across a plurality of data parties, comprising assigning each data party with a corresponding namespace in an object store, assigning a shared namespace in the object store, and triggering a round of federated learning by issuing a customized learning request to at least one data party. Each customized learning request issued to a data party triggers the data party to locally train a model based on training data owned by the data party and model parameters stored in the shared namespace, and upload a local model resulting from the local training to a corresponding namespace in the object store the data party is assigned with. The method further comprises retrieving, from the object store, local models uploaded to the object store during the round of federated learning, and aggregating the local models to obtain a shared model.

Automatically determining poisonous attacks on neural networks

Embodiments relate to a system, program product, and method for automatically determining which activation data points in a neural model have been poisoned to erroneously indicate association with a particular label or labels. A neural network is trained using potentially poisoned training data. Each of the training data points is classified using the network to retain the activations of the last hidden layer, and segment those activations by the label of corresponding training data. Clustering is applied to the retained activations of each segment, and a cluster assessment is conducted for each cluster associated with each label to distinguish clusters with potentially poisoned activations from clusters populated with legitimate activations. The assessment includes executing a set of analyses and integrating the results of the analyses into a determination as to whether a training data set is poisonous based on determining if resultant activation clusters are poisoned.

Calculating relationship strength using an activity-based distributed graph

Methods, systems, and devices for analyzing communication messages (e.g., emails or activities) to determine relationship strength using a distributed graph.