Abstract: Systems and techniques for generating and animating non-player characters (NPCs) within virtual digital environments are provided. Multimodal input data is received that comprises a plurality of input modalities for interaction with an NPC having a set of body features and a set of facial features. The multimodal input data is processed through one or more neural networks to generate animation sequences for both the body features and facial features of the NPC. Generating such animation sequences includes disentangling the multimodal input data to generate substantially disentangled latent representations, combining these representations with the multimodal input data, and using a large-language model (LLM) to generate speech data for the NPC. Further processing using reverse diffusion generates face vertex displacement data and joint trajectory data based on the combined representation and generated speech data.

Abstract: Systems and techniques for generating and animating non-player characters (NPCs) within virtual digital environments are provided. Multimodal input data is received that comprises a plurality of input modalities for interaction with an NPC having a set of body features and a set of facial features. The multimodal input data is processed through one or more neural networks to generate animation sequences for both the body features and facial features of the NPC. Generating such animation sequences includes disentangling the multimodal input data to generate substantially disentangled latent representations, combining these representations with the multimodal input data, and using a large-language model (LLM) to generate speech data for the NPC. Further processing using reverse diffusion generates face vertex displacement data and joint trajectory data based on the combined representation and generated speech data.

Abstract: Systems and techniques for generating and animating non-player characters (NPCs) within virtual digital environments are provided. Multimodal input data is received that comprises a plurality of input modalities for interaction with an NPC having a set of body features and a set of facial features. The multimodal input data is processed through one or more neural networks to generate animation sequences for both the body features and facial features of the NPC. Generating such animation sequences includes disentangling the multimodal input data to generate substantially disentangled latent representations, combining these representations with the multimodal input data, and using a large-language model (LLM) to generate speech data for the NPC. Further processing using reverse diffusion generates face vertex displacement data and joint trajectory data based on the combined representation and generated speech data.

Abstract: An electronic device, a method, and a computer program product generate push notifications directing optimal time for surfacing notifications at a consumer device. The method includes selecting content for consumption at a second device communicatively coupled to a computing device, and encoding into metadata associated with the content, content consumption criteria related to at least one of a network quality and a second device capability for presenting the content at the second device. The method includes transmitting a notification to the second device, the notification including the metadata associated with the content, the metadata configuring the second device to determine at least one of a timing for presenting the content for consumption at the second device and presentation parameters related to the content based on the content consumption criteria.

Abstract: Techniques for object tracking via glasses device are described and are implementable to enable different attributes of objects to be tracked via a glasses device. In implementations, user engagement with an object of interest is detected and tracked. Based on the user engagement with the object of interest, engagement data is generated that includes attributes of the user engagement. Further, cost tracking for objects can be implemented based on images of objects captured via a glasses device. Object records can be created for detected objects and can include object attributes such as object identifiers, object costs, object locations, etc.

Abstract: An electronic device, a method and a computer program product for surfacing notifications based on optimal network and device conditions at a consumer device. The method includes receiving from a network connected device, a notification associated with local consumption of content available for presentation by a communication device, the notification including metadata associated with the content, the metadata encoding content consumption criteria related to preferences for when and how to present the content, in part based on network conditions at the communication device. The method includes detecting one or more current network conditions and comparing the current network conditions with network constraints within the content consumption criteria encoded in the metadata. The method includes selectively presenting the content for consumption at one or more of a time or in a format determined, at least in part, by the detected one or more network conditions.

Abstract: A communication device, a method, and a computer program product for surfacing notifications at a second display of a connected second device based on network conditions and device characteristics at a consumer device and connected second device. The method includes receiving, by a communication device, a notification associated with consumption of content available for presentation by the communication device that interfaces with another electronic device that includes an auxiliary display capable of mirroring a user interface of the communication device. The method includes detecting one or more network conditions and device resource characteristics corresponding to the content consumption criteria encoded in the metadata. The method includes selectively enabling presentation of the content on the auxiliary display for consumption based, at least in part, by the detected one or more network conditions and device resource characteristics.

Abstract: An electronic device, a method, and a computer program product generate push notifications directing optimal time for surfacing notifications at a consumer device. The method includes selecting content for consumption at a second device communicatively coupled to a computing device, and encoding into metadata associated with the content, content consumption criteria related to at least one of a network quality and a second device capability for presenting the content at the second device. The method includes transmitting a notification to the second device, the notification including the metadata associated with the content, the metadata configuring the second device to determine at least one of a timing for presenting the content for consumption at the second device and presentation parameters related to the content based on the content consumption criteria.

Abstract: Provided is a system and method for training and validating models in a machine learning pipeline for failure mode analytics. The machine learning pipeline may include an unsupervised training phase, a validation phase and a supervised training and scoring phase. In one example, the method may include receiving a request to create a machine learning model for failure mode detection associated with an asset, retrieving historical notification data of the asset, generating an unsupervised machine learning model via unsupervised learning on the historical notification data, wherein the unsupervised learning comprises identifying failure topics from text included in the historical notification data and mapping the identified failure topics to a plurality of predefined failure modes for the asset, and storing the generated unsupervised machine learning model via a storage device.

Abstract: A method of assembling a pipeline at a seabed location comprises landing a connection tool on the seabed over a free end portion of a first pipeline section already placed on the seabed. The connection tool is locked to the free end portion of the first pipeline section, a lower end of a second pipeline section is connected to the connection tool via an initiation line. While applying tension to the initiation line against reaction force of the connection tool, at least a lower end portion of the second pipeline section is landed on the seabed with the lower end facing a free end of the first pipeline section. The lower end of the second pipeline section is then pulled into mechanical engagement with the free end of the first pipeline section.

Abstract: A method of installing a subsea pipeline supports at least one elongate pipe stalk at the surface of the sea at a first, relatively shallow-water location by virtue of buoyancy added to the pipe stalk. The pipe stalk is then towed at the surface to a second location that is in deeper water. There, with the pipe stalk supported between leading and trailing towing vessels, at least some of the added buoyancy is removed. This causes the pipe stalk to hang with catenary curvature beneath the surface between the vessels. The catenary-curved pipe stalk hanging between the vessels is then towed to a third location for subsea installation, which may involve upending the pipe stalk before landing a lower end portion of it on the seabed.

Abstract: Provided is a system and method for training and validating models in a machine learning pipeline for failure mode analytics. The machine learning pipeline may include an unsupervised training phase, a validation phase and a supervised training and scoring phase. In one example, the method may include receiving a request to create a machine learning model for failure mode detection associated with an asset, retrieving historical notification data of the asset, generating an unsupervised machine learning model via unsupervised learning on the historical notification data, wherein the unsupervised learning comprises identifying failure topics from text included in the historical notification data and mapping the identified failure topics to a plurality of predefined failure modes for the asset, and storing the generated unsupervised machine learning model via a storage device.

Abstract: Systems, devices, and techniques are disclosed for multi-threaded dynamic queries on an unpartitioned database. An instruction to perform a database operation on rows of a database may be received. Processing threads may be generated by generating three random numbers. A starting point of a processing thread may be set to a first random number, an offset may be set to a second random number, and an order may be based on a third random number. The processing threads may each generate a dynamic query. A dynamic query generated by a processing thread may include an instruction to perform the database operation on rows of the database based on the offset, starting point, and order for the processing thread. Each dynamic query may be run against the database to perform the database operation.

Abstract: Examples include a ticketing system for managing performance of computer services. The ticketing system includes a ticket database to store a plurality of tickets, each ticket defining a job to be performed on a computing system, and a master to create a ticket, store the ticket in the ticket database, and update the ticket. The ticketing system includes a scheduler to determine whether to create the ticket based at least in part on a set of conditions, and to request the master to create the ticket when the ticket is to be created, and a worker to get the ticket from the master, perform the job defined by the ticket, and request the master to update the ticket with a status of the ticket. The ticketing system includes a supervisor to request the master to find a stale ticket, and, if found, to request the worker to abort performance of the job defined by the stale ticket.

Abstract: Provided is a system and method for training and validating models in a machine learning pipeline for failure mode analytics. The machine learning pipeline may include an unsupervised training phase, a validation phase and a supervised training and scoring phase. In one example, the method may include receiving a request to create a machine learning model for failure mode detection associated with an asset, retrieving historical notification data of the asset, generating an unsupervised machine learning model via unsupervised learning on the historical notification data, wherein the unsupervised learning comprises identifying failure topics from text included in the historical notification data and mapping the identified failure topics to a plurality of predefined failure modes for the asset, and storing the generated unsupervised machine learning model via a storage device.

Abstract: A method of assembling a pipeline at a seabed location comprises landing a connection tool (10) on the seabed over a free end portion of a first pipeline section (12) already placed on the seabed. The connection tool is locked to the free end portion of the first pipeline section, a lower end of a second pipeline section (26) is connected to the connection tool via an initiation line (68). While applying tension to the initiation line against reaction force of the connection tool, at least a lower end portion of the second pipeline section is landed on the seabed with the lower end facing a free end of the first pipeline section. The lower end of the second pipeline section is then pulled into mechanical engagement with the free end of the first pipeline section.

Abstract: Provided is a system and method for predicting leading indicators for predicting occurrence of an event at a target asset. Rather than rely on traditional manufacturer-defined leading indicators for an asset, the examples herein predict leading indicators for a target asset based on actual operating conditions at the target asset. Accordingly, unanticipated operating conditions can be considered. In one example, the method may include receiving operating data of a target resource, the operating data being associated with previous occurrences of an event at the target resource, predicting one or more leading indicators of the event at the target resource based on the received operating data, each leading indicator comprising a variable and a threshold value for the variable, and outputting information about the one or more predicted leading indicators of the target resource for display via a user interface.

Abstract: A method of installing a subsea pipeline supports at least one elongate pipe stalk at the surface of the sea at a first, relatively shallow-water location by virtue of buoyancy added to the pipe stalk. The pipe stalk is then towed at the surface to a second location that is in deeper water. There, with the pipe stalk supported between leading and trailing towing vessels, at least some of the added buoyancy is removed. This causes the pipe stalk to hang with catenary curvature beneath the surface between the vessels. The catenary-curved pipe stalk hanging between the vessels is then towed to a third location for subsea installation, which may involve upending the pipe stalk before landing a lower end portion of it on the seabed.

Abstract: Examples include a computer-implemented method for determining one or more observations about a computing system, determining a plurality of sets of one or more actions based at least in part on a plurality of policies and the one or more observations; reducing the plurality of sets of one or more actions to generate a reduced set of actions, the reduced set of actions including scheduling a computing system to execute the reduced set of actions by requesting one or more idempotent tickets to be created in a ticketing system; and performing, or causing to be performed, the reduced set of actions.

Abstract: System and methods are described for restoring a database by receiving a request to restore a database in a cloud computing environment, generating a restore plan for the database based at least in part on previously stored database restore outcomes and error probabilities of full backups, incremental backups, and redo logs, restoring the database from an archive according to the restore plan, and storing an outcome of the database restore.