Abstract: Systems and methods for optimizing key performance indicators (KPIs) using adversarial imitation deep learning include processing sensor data received from sensors to remove irrelevant data based on correlation to a final KPI and generating, using a policy generator network with a transformer-based architecture, an optimal sequence of actions based on the processed sensor data. A discriminator network is employed to differentiate between the generated action sequences and real-world high performance sequences employing. Final KPI results are estimated based on the generated action sequences using a performance prediction network. The generated action sequences are applied to the process to optimize the KPI in real-time.

Abstract: Systems and methods train a transformer-based policy network and Generative Adversarial Network (GAN) by initializing a transformer-based policy network to model action sequences by encoding temporal dependencies within sensor data. Multi-head self-attention mechanisms process sequential sensor inputs by being pre-trained on a labeled dataset having sensor data from known low-risk action sequences. A generator within the GAN is trained to produce generated action sequences, which mimic behavior of low-risk action sequences. A discriminator within the GAN is concurrently trained to differentiate between action sequences derived from the labeled dataset and synthetic action sequences produced by the generator. A feedback loop is employed to adjust parameters to produce sequences indistinguishable from real low-risk action sequences.

Abstract: Systems and methods for an agent-based carbon emission reduction system. A carbon product of a supply chain system can be limited below a carbon product threshold by performing a corrective action to monitored entities based on a calculated carbon emission. The carbon emission can be calculated based on carbon-relevant data and a calculation route by utilizing an agent-based simulation model that simulates a learned relationship between a supply chain system and the carbon-relevant data. The calculation route can be determined based on the carbon-relevant data based on a relevance of a carbon product contribution of monitored entities to a goal of the monitored entities. Carbon-relevant data can be extracted from the monitored entities.

Abstract: Systems and methods are provided for classifying components include monitoring sensors to collect sensor data related to a state of a plurality of components; processing, by a computing system, the sensor data to generate an action sequence using a transformer-based policy network for each of the components. A risk score is generated for the action sequence using a Generative Adversarial Network (GAN), wherein the GAN includes a generator for generating action sequences and a discriminator to distinguish low-risk action sequences in accordance with a threshold. The low-risk action sequences are associated with components in the plurality of components based on the risk score. A status of the low-risk action sequences is communicated to the components.

Abstract: A distributed optimization control method for an aircraft energy system considering loss includes the following steps: S1, based on characteristics of the actual near-space aircraft energy system and the near-space environment, the multi-bus DC microgrid system model is established for the near-space aircraft energy system according to Kirchhoff's law; S2, according to the energy scheduling requirement of the near-space aircraft energy system, the system control target is established; S3, according to the system control target, the optimal scheduling problem of the aircraft energy system considering operating loss is established; and S4, based on projection and penalty, the distributed optimization control algorithm is designed. The distributed optimization control method realizes bus voltage regulation and current distribution.

Abstract: Systems and methods for a demonstration uncertainty-based artificial intelligence model for open information extraction. A large language model (LLM) can generate initial structured sentences using an initial prompt for a domain-specific instruction extracted from an unstructured text input. Structural similarities between the initial structured sentences and sentences from a training dataset can be determined to obtain structurally similar sentences. The LLM can identify relational triplets from combinations of tokens from generated sentences using and the structurally similar sentences. The relational triplets can be filtered based on a calculated demonstration uncertainty to obtain a filtered triplet list. A domain-specific task can be performed using the filtered triplet list to assist the decision-making process of a decision-making entity.

Abstract: Systems and methods for correlation-aware explainable online change point detection. Collected data metrics from the cloud system can be transformed to correlation matrices. Correlation shifts from the correlation matrices can be captured as differences of correlation between batches of collected data metrics through determined statistics of the batches of collected data metrics across timesteps. Change points in the cloud system can be detected based on the correlation shifts to obtain detected change points. System maintenance can be performed autonomously based on the detected change points from identified system entities to optimize the cloud system with an updated configuration.

Abstract: A method for determining an optimal sensor set includes identifying a set of possible faults of a thermal hydraulic system and a set of diagnostic objectives. The method also includes obtaining descriptions of sensor sets, which includes: receiving a first description of a first sensor set, and generating, based on the first description, a second description of a second sensor set. The method further includes, for each sensor set: determining a diagnostic capability of the sensor set, and calculating a score of the particular sensor set based on the particular description and the diagnostic capability, wherein a score of a sensor set is increased by a monetary cost of the sensor set and decreased by the sensor set meeting a diagnostic objective. The method also includes identifying the optimal sensor set that has a lowest score of the sensor sets, and displaying an indication of the optimal sensor set.

Abstract: A steering wheel instrument assembly includes a button assembly having a microprocessor and a push-button switch. A function table has a unified format code associated with the push-button switch and indicates a desired vehicle operation. The assembly includes a vehicle protocol adapter module having another microprocessor with a second function table having the unified format code associated with an OEM format code. The microprocessor of the button assembly retrieves the unified format code when the push-button switch is activated and sends the unified format code to the vehicle protocol adapter module. The vehicle protocol adapter module retrieves the OEM format code from the second function table in response to receiving the unified format code and sends the OEM format code to an engine control unit.

Abstract: This disclosure proposes systems and methods for entity resolution using speech recognition data. The system can receive audio data representing an utterance and perform automatic speech recognition (ASR) processing on the audio data to generate at least a first ASR data and a second ASR data. The system can perform natural language understanding (NLU) processing on the first ASR data to determine intent data and an indication of an entity. The system can determine a first portion of the first ASR data that corresponds to the indication of the entity. The system can determine a second portion of the second ASR data that corresponds to the indication of the entity without performing NLU on the second ASR data. The system can perform entity resolution (ER) on the second portion to identify a first entity.

Abstract: Methods and systems for information extraction include configuring a language model with an information extraction instruction prompt and at least one labeled example prompt. Configuration of the language model is validated using at least one validation prompt. Errors made by the language model in response to the at least one validation prompt are corrected using a correction prompt. Information extraction is performed on an unlabeled sentence using the language model to identify a relation from the unlabeled sentence. An action is performed responsive to the identified relation.

Abstract: A handheld power tool includes a motor, a handle having a first switch for starting and stopping the motor, an output assembly, and an adjustment device including a telescopic rod assembly. The length of the adjustment device changes so that the handheld power tool has at least first and second working states. The distance between the frontmost side of the handheld power tool and the rearmost side of the first switch is an effective length of the handheld power tool. When the handheld power tool is in the first working state, the handheld power tool has a first effective length T1. When the handheld power tool is in the second working state, the handheld power tool has a second effective length T2. The first effective length T1 is less than or equal to 1 m, and the second effective length T2 is greater than the first effective length T1.

Abstract: The present disclosure relates to a PointEFF method for urban object classification with LiDAR point cloud data, and belongs to the field of LiDAR point cloud classification. The method comprises: point cloud data segmentation; End-to-end feature extraction layer construction; External feature fusion layer construction; and precision evaluation.

Abstract: The present disclosure provides an information processing method, an apparatus, an electronic device, and a storage medium. By establishing a session group including a first identity user and a second identity user in response to the first identity user satisfying a preset first condition, pushing a first session message for the first identity user within the session group according to a preset process node, and displaying a first information editing interface for the first identity user, in response to a first preset operation of the first identity user for the first session message, the present disclosure can achieve automatic group creation, automatic reminder through group session, and entry into a relevant information editing interface through group session, thereby achieving efficient distribution and processing of information.

Abstract: The present disclosure relates to a method and system for simulating water-induced rock strength deterioration based on a discrete element method, and relates to the field of simulation of water-induced rock strength deterioration.

Abstract: A pole chain saw includes a connecting rod and a front end assembly disposed at the front end of the connecting rod. The front end assembly includes an electric motor; an output device including a guide plate and a chain wound around the guide plate, where the electric motor drives the chain to rotate; a transmission assembly connecting the electric motor to the output device; and a gearbox supporting and accommodating the transmission assembly, where at least part of the gearbox is made of metal. The chain saw can effectively enhance the heat dissipation effect of the gearbox and increase the service life of a transmission mechanism.

Abstract: A method for cultivating a Tung tree seedling through hypocotyle grafting includes selection of a sand storage time of a rootstock (Vernicia montana Lour.) seed, seedbed construction, scion collection, scion cutting, rootstock cutting, bandaging, transplantation and management, and the like. A grafted plant obtained by a traditional grafting method (bud grafting) for Tung tree is easy to break at a grafting interface; and when the stem grafting is conducted, because a scion is thick, a rootstock for grafting must be cultivated one year in advance. These grafting methods have production problems such as long cycle and high cost, which can be solved by the present disclosure. The present disclosure has a short rootstock cultivation time, simple grafting operations, a high grafting efficiency, and a high survival rate, greatly shortens a cycle of asexual propagation-based seedling cultivation for Tung tree.

Abstract: An artificial intelligence (AI) interaction method, system, and computer program product include selecting an artificial intelligence model to respond to a query to generating a response to the query using the selected artificial intelligence model, and receiving the response to the query from the selected artificial intelligence model.

Abstract: A handheld cutting tool includes a housing, a motor, a saw chain, a guide plate, and a blade sheath, where the motor is supported by the housing, the saw chain is driven by the motor to perform a cutting function, an end of the guide plate is supported by the housing and the guide plate is used for supporting and guiding the saw chain, and the blade sheath is formed with or connected to the first limiting portion, the housing or the guide plate is formed with a second limiting portion, and the first limiting portion and the second limiting portion are adaptable to each other so that the blade sheath is not separated from the handheld cutting tool at least under the action of gravity.