Abstract: Methods and apparatus are disclosed for customizing a copilot. Document records related to copilot objectives or tasks are obtained and used to identify corresponding data sources. Data sources can be integrated into data producers or data repositories, to be used by a retrieval microservice. Data producers and other microservices are individually fine-tuned for the custom application, before or after integration into the copilot. The integrated copilot is tested end-to-end, and can be further refined. Disclosed techniques range from fully automated to human-in-the-loop (e.g. guided by an expert) to fully interactive. Some techniques produce tools which can automate certain customization operations. Training data, tasks, or document records blend expert-generated, developer-generated, or synthesized items.

Abstract: Methods and apparatus are disclosed for customizing a copilot or other machine learning tool. Document records related to copilot objectives or tasks are obtained and used to identify corresponding data sources. Data sources can be integrated into data producers or data repositories, to be used by a retrieval microservice. Data producers and other microservices are individually fine-tuned for the custom application, before or after integration into the copilot. The integrated copilot is tested end-to-end, and can be further refined. Disclosed techniques range from fully automated to human-in-the-loop (e.g. guided by an expert) to fully interactive. Some techniques produce tools which can automate certain customization operations. Training data, tasks, or document records blend expert-generated, developer-generated, or synthesized items.

Abstract: Apparatus and methods are disclosed for implementing a copilot as a network of microservices including specialized large language models (LLMs) or other trained machine learning (ML) tools. This architecture supports flexible, customizable, or dynamically determinable dataflow. Compared to much larger competing LLMs, comparable or superior performance is achieved, while significantly reducing computation time and hardware requirements, even to a single compute node with a single GPU. Examples incorporate a retrieval microservice, as least one data producer, and a core microservice. Based on client input, the retrieval microservice can perform multiple iterations of retrieval augmented generation (RAG). At each iteration, output (based on any preceding iterations' results or the client input) is transmitted to a data producer, and results received therefrom. Eventually, based on these results, an output is transmitted toward the core microservice for generation of a response to the client input.

Abstract: Apparatus and methods are disclosed for generating matching application programming interface (API) queries for a received input. A matching score is generated between each of a group of candidate API queries and the received input. One or more matching API queries are selected from the group based on their respective matching scores. Disclosed techniques are suitable for a data producer front end in a copilot having a microservice network architecture. This architecture supports flexible, customizable, or dynamically determinable dataflow. Compared to much larger competing LLMs, comparable or superior performance is achieved for certain tasks, while computation time and hardware requirements are significantly reduced, even to a single compute node with a single GPU. One or more data producers can provide a retrieval microservice with access to various databases having respective APIs, to extend the copilot's reach. Variations and additional techniques are disclosed.

Abstract: Apparatus and methods are disclosed for implementing a copilot as a network of microservices including specialized large language models (LLMs) or other trained machine learning (ML) tools. This architecture supports flexible, customizable, or dynamically determinable dataflow. Compared to much larger competing LLMs, comparable or superior performance is achieved for certain tasks, while significantly reducing computation time and hardware requirements, even to a single compute node with a single GPU. An expansion microservice converts client input tokens into associated tokens to diversify targeted tasks presented to a core microservice. An ML tool in the expansion microservice is pretrained in multiple stages for varying tasks using varying general and target-specific corpora. Synthesized training data derived from a knowledge graph can also be used. The expansion ML tool is subsequently fine-tuned in one or multiple stages. Variations and additional techniques are disclosed.

Abstract: Methods and apparatus are disclosed for customizing a copilot or other machine learning tool. Document records related to copilot objectives or tasks are obtained and used to identify corresponding data sources. Data sources can be integrated into data producers or data repositories, to be used by a retrieval microservice. Data producers and other microservices are individually fine-tuned for the custom application, before or after integration into the copilot. The integrated copilot is tested end-to-end, and can be further refined. Disclosed techniques range from fully automated to human-in-the-loop (e.g. guided by an expert) to fully interactive. Some techniques produce tools which can automate certain customization operations. Training data, tasks, or document records blend expert-generated, developer-generated, or synthesized items.

Abstract: Apparatus and methods are disclosed for implementing a copilot as a network of microservices including specialized large language models (LLMs) or other trained machine learning (ML) tools. The microservice network architecture supports flexible, customizable, or dynamically determinable dataflow. Compared to much larger competing LLMs, comparable or superior performance is achieved for certain tasks, while significantly reducing computation time and hardware requirements, even to a single compute node with a single GPU. Examples incorporate a qualification microservice to test data, destined for a downstream microservice, for conformance with the copilot's competency. A knowledge graph of a target domain is built, visualized, and pruned. The data is tested for conformance with the pruned domain representation and non-conforming data is discarded. Variations and additional techniques are disclosed.

Abstract: Apparatus and methods are disclosed for implementing a copilot as a network of microservices including specialized large language models (LLMs) or other trained machine learning (ML) tools. The microservice network architecture supports flexible, customizable, or dynamically determinable dataflow. Compared to much larger competing LLMs, comparable or superior performance is achieved for certain tasks, while significantly reducing hardware requirements and computation time. Disclosed examples incorporate microservices for expansion, retrieval, embedding, and evaluation, in addition to one or more core microservices. Optionally, intermodal I/O, multiple data repositories, competency qualification, or human feedback can be supported. Multiple core microservices can support varying client authorizations or cognitive functions. The disclosed architecture supports any major LLM use case and can be deployed on a single compute node with a single GPU.

Abstract: Recorded sessions of skilled personnel at work are used to train a first machine learning (ML) tool to extract workflow maps. These maps are used to train a second ML tool to emulate at least one workflow. The first ML tool is trained to predict an annotator's output. Either ML tool can be a copilot having a microservice network architecture. Further, complex sets of workflows can be subdivided for efficient support with small ML tools for (1) workflow map extraction and (2) emulation. Based on segment demarcation accompanying a recording, segments are assigned to specialized ML extraction tools for workflow map extraction. Extracted workflow maps are used to train workflow emulator(s). Similarly, specialized ML emulation tools can emulate respective tasks. A distribution microservice can identify a task to be emulated and can invoke the appropriate specialized ML tool. Similar microservice network architectures support both applications.

Abstract: Methods and apparatus are disclosed for customizing a copilot. Document records related to copilot objectives or tasks are obtained and used to identify corresponding data sources. Data sources can be integrated into data producers or data repositories, to be used by a retrieval microservice. Data producers and other microservices are individually fine-tuned for the custom application, before or after integration into the copilot. The integrated copilot is tested end-to-end, and can be further refined. Disclosed techniques range from fully automated to human-in-the-loop (e.g. guided by an expert) to fully interactive. Some techniques produce tools which can automate certain customization operations. Training data, tasks, or document records blend expert-generated, developer-generated, or synthesized items.

Abstract: Systems and methods are provided for generating a statistical parameter representing a state of a surgical procedure from sensor data. Sensor data representing a time period. is received from a sensor. Numerical features representing the time period are generated from the sensor data. Each of a plurality of long short term memory units are updated according to the plurality of numerical features via a message passing process. The long short term memory units are connected to form a graph, with a first set of the long short term memory units representing a plurality of nodes of the graph and a second set of the long short term memory units representing a plurality of hyperedges of the graph. A statistical parameter representing a state of the surgical procedure for the time period is derived from an output of one of the long short term memory units and provided to a user.

Abstract: Methods and systems for transmitting an interruption-free version of a television program for recording on a user equipment device. In one aspect, a system transmits a first stream including a television program to a user equipment device. The system receives an indication that the television program is being recorded on the user equipment device. The system determines whether an interruption is imminent during transmission of the television program. In response to determining that the interruption is imminent, the system generates a second stream including the television program. The second stream does not include the interruption during the transmission of the television program. The system transmits the second stream to the user equipment device along with an instruction to switch recording of the television program from the first stream to the second stream.

Abstract: Systems and methods are described herein for presenting additional information related to an emergency alert. A media guidance application may detect that a new emergency alert has been received. The media guidance application may then perform a search of various content sources (e.g., program listings, the Internet, etc.) for content related to the emergency alert. The media guidance application may then present the search results in a contextual menu enabling the user to access content of the search results. The media guidance application may also provide more options, such as, recording or pausing media content that the user was in the process of consuming when the emergency alert was generated, and an option to the user not to bring up the contextual menu again if the same emergency alert is repeated.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for training an action selection neural network used to select actions to be performed by an agent interacting with an environment. In one aspect, a system comprises a plurality of actor computing units and a plurality of learner computing units. The actor computing units generate experience tuple trajectories that are used by the learner computing units to update learner action selection neural network parameters using a reinforcement learning technique. The reinforcement learning technique may be an off-policy actor critic reinforcement learning technique.

Abstract: The End-to-End Encryption and Hot Wallet Key Recovery Apparatuses, Processes and Systems (“E2EEHWKR”) transforms key backup request, key recovery request datastructure/inputs via E2EEHWKR components into key backup response, key recovery response outputs. A key backup request datastructure specifying a wallet private key, a private key backup server identifier, a PIN shard fracture scheme definition, and a plurality of PIN shard backup devices is obtained. A user security PIN is obtained and utilized to encrypt the wallet private key. A symmetric key is calculated from a public key associated with the private key backup server identifier and an app private key. The encrypted wallet private key is encrypted utilizing the symmetric key. The twice encrypted wallet private key is sent to a private key backup server. The user security PIN is encrypted utilizing an asymmetric keypair. Encrypted user security PIN shards are generated and sent to PIN shard backup devices.

Abstract: A gas hydrate inhibitor composition is disclosed including two main components: (1) 5-95% of an amphiphilic compound with a hydrophobic tail linked to a hydrophilic head group through various linking structures, and (2) 5-95% of a cationic surfactant based on di(C8-C18 alkyl)dimethyl ammonium salts. The linking structures include amide-based linkers and more complex structures containing hydroxyl groups, carboxylic acid groups, and various combinations. The composition features specific requirements for substituent groups ranging from small alkyl chains to larger hydrocarbyl groups.

Abstract: Systems, methods, and computer-readable storage devices for providing directions haptically such that sight and hearing can continue unimpeded. In one exemplary embodiment, a wearable device (such as earphones, ear rings, gloves, glasses, or other wearable objects) configured as disclosed herein receives directions to an intended destination for a user, the directions comprising a movement action and a distance to the movement action. The wearable device has multiple haptic output units and generates, through one of those units, a haptic output based on the directions. This allows the user to receive the directions through touch rather than looking at their mobile device or from audio.

Abstract: Apparatus and methods are disclosed for implementing a copilot as a network of microservices including specialized large language models (LLMs) or other trained machine learning (ML) tools. The microservice network architecture supports flexible, customizable, or dynamically determinable dataflow. Compared to much larger competing LLMs, comparable or superior performance is achieved for certain tasks, while significantly reducing hardware requirements and computation time. Disclosed examples incorporate microservices for expansion, retrieval, embedding, and evaluation, in addition to one or more core microservices. Optionally, intermodal I/O, multiple data repositories, competency qualification, or human feedback can be supported. Multiple core microservices can support varying client authorizations or cognitive functions. The disclosed architecture supports any major LLM use case and can be deployed on a single compute node with a single GPU.

Abstract: This disclosure describes systems, methods, and apparatus for a locking folding knife having a handle lock and a safety switch arranged around and rotatable around a pivot axis of the knife. In an engaged position the switch overlaps at least a portion of the handle lock and prevents it from moving into an unlocked position. A spring, such as a wire form spring, can be arranged within a pocket of the switch, and when the switch is moved between engaged and disengaged positions the spring can be compressed generating a rotational bias on the switch urging the switch toward either the engaged or disengaged position.

Abstract: Systems and methods are described herein for presenting additional information related to an emergency alert. A media guidance application may detect that a new emergency alert has been received. The media guidance application may then perform a search of various content sources (e.g., program listings, the Internet, etc.) for content related to the emergency alert. The media guidance application may then present the search results in a contextual menu enabling the user to access content of the search results. The media guidance application may also provide more options, such as, recording or pausing media content that the user was in the process of consuming when the emergency alert was generated, and an option to the user not to bring up the contextual menu again if the same emergency alert is repeated.