Abstract: The present disclosure provides a method of preparing a radio frequency filter, including: a substrate; a supporting electrode protruded on a front surface of the substrate; and a thin film structure formed on the substrate and spaced with the substrate by the supporting electrode. An end surface of a top end of the supporting electrode is in sealing contact with a front surface of the thin film structure.

Abstract: A method for predicting and optimizing an ROP for oil and gas drilling based on Bayesian optimization includes: acquiring raw drilling data according to a preset sampling period, constructing an initial sample data set based on the raw drilling data, constructing an ROP prediction model based on the initial sample data set, and predicting an ROP at a next sample point through a Gaussian process regression based on the ROP prediction model. The present invention realizes rapid analysis of historical drilling data and accurate prediction of an ROP range at a sample point in a feasible domain. The method can obtain an optimized ROP and an optimized engineering parameter through Bayesian optimization, and obtain an engineering parameter corresponding to an optimal ROP. The method has few restrictions on the drilling engineering parameter and the raw formation parameter, improves prediction accuracy, and avoids the problem of blurred parameter value boundaries.

Abstract: Leveraging domain knowledge is an effective strategy for enhancing the quality of inferred low-dimensional representations of documents by topic models. Presented herein are embodiments of a Bayesian nonparametric model that employ knowledge graph (KG) embedding in the context of topic modeling for extracting more coherent topics; embodiments of the model may be referred to as topic modeling with knowledge graph embedding (TMKGE). TMKGE embodiments are hierarchical Dirichlet process (HDP)-based models that flexibly borrow information from a KG to improve the interpretability of topics. Also, embodiments of a new, efficient online variational inference method based on a stick-breaking construction of HDP were developed for TMKGE models, making TMKGE suitable for large document corpora and KGs. Experiments on datasets illustrate the superior performance of TMKGE in terms of topic coherence and document classification accuracy, compared to state-of-the-art topic modeling methods.

Abstract: Provided are a solidly mounted resonator and a method for preparing a solidly mounted resonator. The solidly mounted resonator includes: a piezoelectric structure, wherein the piezoelectric structure includes: an upper electrode layer, a lower electrode layer and a piezoelectric layer. The lower electrode layer disposed corresponding to the piezoelectric structure, and the lower electrode layer includes: a protruding portion protruding downward corresponding to a lower surface of the lower electrode layer; the piezoelectric layer is disposed on an upper surface of the lower electrode layer; the upper electrode layer is disposed on an upper surface of the piezoelectric layer.

Abstract: Presented herein are embodiments of a bi-level optimization framework an inner loop phase optimizes an example-level problem to generate robust exemplars, while an outer loop phase proposes an adaptive optimization to achieve the robustness of the projected DNN models. Embodiments for watermarking a deep neural network include obtaining a set of temporary parameters for a temporary model. The set of temporary parameters may be generated based upon a set of base parameters of a base model. Embodiments may further include generating a set of boundary watermark exemplars using the set of temporary parameters for the temporary model. In one or more embodiments, the set of boundary watermark exemplars maximizes an identification loss of the temporary model on a set of watermark data. Embodiments may further include outputting a watermark embedded base model by embedding the set of boundary watermark exemplars into one or more base parameters of the base model.

Abstract: A pharmaceutical composition for treating diseases related to inflammation or injury such as ARDS, comprising: (a) an active material derived from human somatic cells, the active ingredient being an extracellular vesicle produced by human somatic cells; and (b) a pharmaceutically acceptable carrier. The somatic cells are selected from human tissue, bone marrow and/or blood-derived stem cells, etc. The dosage form of the pharmaceutical composition is selected from an atomized inhalation agent, eye drops and nasal drops. The pharmaceutical composition has storage stability and high dispersibility superior to those of a living cell formulation. A method for preparing an extracellular vesicle, comprising steps of culturing human somatic cells, removing cells from a culture system, mixing a culture solution with polyethylene glycol to form an extracellular vesicle modified by PEG, centrifuging, resuspending, etc.

Abstract: Described herein are embodiments for end-to-end reinforcement learning based coreference resolution models to directly optimize coreference evaluation metrics. Embodiments of a reinforced policy gradient model are disclosed to incorporate reward associated with a sequence of coreference linking actions. Furthermore, maximum entropy regularization may be used for adequate exploration to prevent a model embodiment from prematurely converging to a bad local optimum. Experiments on datasets compared with state-of-the-art methods verified the effectiveness of embodiments.

Abstract: This disclosure provides systems, methods, and devices for wireless communication that support enhanced beam management using extended reality (XR) perception data. In a first aspect, a method of wireless communication includes establishing a communication connection between a user equipment (UE) and a serving base station using a current serving beam selected by the UE from a plurality of available beams paired with a serving base station beam. The method further includes obtaining, perception information from one or more extended reality sensors associated with the UE and determining, in response to detection of UE movement, a transpositional representation of the movement using the perception information. The UE may then select a new serving beam in accordance with the transpositional representation. Other aspects and features are also claimed and described.

Abstract: A method for generating a model includes: obtaining training data for training a target model, in which the training data includes labeled data and unlabeled data; obtaining a first output result by the target model based on the labeled data; obtaining a second output result by the target model based on the unlabeled data; and obtaining an optimized target model by optimizing the target model based on the first output result and the second output result.

Abstract: A dual-frequency low-noise amplifier circuit includes an input impedance matching circuit, a radio-frequency signal amplifying circuit and an output impedance matching circuit. An input end of the input impedance matching circuit inputs a radio-frequency signal, and an output end is connected to an input end of the output impedance matching circuit by means of the radio-frequency signal amplifying circuit; an output end of the output impedance matching circuit outputs an amplified radio-frequency signal; a first band radio-frequency signal is amplified, input and output switches are in a first open and closed state, and input and output impedances are first input and output impedances; and a second band radio-frequency signal is amplified, the input and output switches are in a second open and closed state, and the input and output impedances are second input and output impedances. Adjusting the dual-frequency low-noise amplifier circuit can reduce signal reflection and improve signal quality.

Abstract: A dual-band low-noise amplifier circuit includes an amplification sub-circuit and a switch frequency selection circuit; the amplification sub-circuit is used for performing gain amplification on a radio frequency signal to be amplified to obtain an amplified radio frequency signal, and outputting the amplified radio frequency signal; the switch frequency selection circuit is connected to the amplification sub-circuit, and is used for controlling the state of a switch in the switch frequency selection circuit on the basis of a target frequency band corresponding to the radio frequency signal to be amplified, so that the dual-band low-noise amplifier circuit meets optimal performance in the target frequency band. In this way, low-noise amplification of dual-band signals is achieved by means of the reconfigurable structure of the low-noise amplifier circuit, and parameters such as noise figure, gain, and linearity can be kept in optimal states in each frequency band.

Abstract: Various aspects of the present disclosure provide for enabling at least one opportunity to transmit mission critical (MiCr) data and at least one opportunity to receive MiCr data in a time division duplex (TDD) subframe during a single transmission time interval (TTI). The single TTI may be no greater than 500 microseconds. The TDD subframe may be a downlink (DL)-centric TDD subframe or an uplink (UL)-centric TDD subframe. How much of the TDD subframe is configured for the at least one opportunity to transmit the MiCr data and how much of the TDD subframe is configured for the at least one opportunity to receive the MiCr data may be adjusted based on one or more characteristics of the MiCr data. The MiCr data may have a low latency requirement, a high priority requirement, and/or a high reliability requirement. Various other aspects are provided throughout the present disclosure.

Abstract: A nano-wall integrated circuit structure with high integration density is disclosed, which relates to the fields of microelectronic technology and integrated circuits (IC). Based on the different device physical principles with MOSFETs in traditional ICs, the nano-wall integrated circuit unit structure (Nano-Wall FET, referred to as NWaFET) with high integration density can improve the integration of the IC, significantly shorten the channel length, improve the flexibility of the device channel width-to-length ratio adjustment, and save chip area.

Abstract: Methods, systems, and devices for wireless communications are described. Generally, the described techniques provide for identification of a slot format for a set of slots where at least one slot format of the set is an subband full-duplex slot (e.g., includes uplink and downlink resources separated in a frequency domain). The base station may transmit one or more semi-persistent scheduling configurations to a UE. The UE may identify semi-persistent scheduling occasions for receiving a downlink communication based at least in part on the SPS configurations matching the format of the slot, the periodicity of the slot formats of the set, or other considerations. The UE and the base station may communicate based at least in part on the determined semi-persistent scheduling occasions.

Abstract: This disclosure provides systems, methods, and devices for wireless communication that support enhanced beam management using extended reality (XR) perception data. In a first aspect, a method of wireless communication includes establishing a communication connection between a user equipment (UE) and a serving base station using a current serving beam selected by the UE from a plurality of available beams paired with a serving base station beam. The method further includes obtaining, perception information from one or more extended reality sensors associated with the UE and determining, in response to detection of UE movement, a transpositional representation of the movement using the perception information. The UE may then select a new serving beam in accordance with the transpositional representation. Other aspects and features are also claimed and described.

Abstract: This invention provides methods for extracting ingredients and materials from plant material, including extracting ingredients from cannabis plant material.

Abstract: Described herein are embodiments of a unified neural network framework to integrate Topic modeling, Word embedding and Entity Embedding (TWEE) for representation learning of inputs. In one or more embodiments, a novel topic sparse autoencoder is introduced to incorporate discriminative topics into the representation learning of the input. Topic distributions of inputs are generated from a global viewpoint and are utilized to enable autoencoder to learn topical representations. A sparsity constraint may be added to ensure that the most discriminative representations are related to topics. In addition, both words and entity related information may be embedded into the network to help learn a more comprehensive input representation. Extensive empirical experiments show that embodiments of the TWEE framework outperform the state-of-the-art methods on different datasets.

Abstract: The present disclosure discloses a split type motor, including an air channel and an armature module, wherein the armature module includes a housing, a stator assembly, a rotor assembly, and a printed circuit board (PCB); the housing is cylindrical; the rotor assembly includes a rotating shaft, a magnetic ring, a first bearing, and a second bearing; the housing is provided with a first bearing chamber and a second bearing chamber; an elastic component is arranged between the first bearing and the magnetic ring; the air channel includes an inner ring and an outer ring concentrically arranged; the inner ring encloses and forms a circular through cavity matched with the housing; a ventilation cavity is formed between the inner ring and the outer ring; and several flow guide plates are arranged in the ventilation cavity.

Abstract: The present invention relates to a shoe upper including a knitted component having a plurality of integrally knitted lines of loops including a set of consecutive lines, wherein each line in the set of consecutive lines includes a plurality of consecutive loops which are interlooped with adjacent loops of the respective line and not interlooped with loops of a line which is not contained in the set, and wherein the set of consecutive lines forms at least a portion of an eyelet which is configured to receive a lace.

Abstract: Methods, systems, and devices for wireless communication are described. In one example, a base station may dynamically configure a user equipment (UE) to monitor or avoid monitoring for a preemption indication. Accordingly, the UE may have a lower chance of monitoring for a preemption indication when it is unlikely that an uplink or downlink transmission will be preempted. In another example, a base station may transmit a control message indicating whether future indications received from the base station are to be interpreted as preemption indications or permission indications. Accordingly, the base station may choose to use either preemption indications or permission indications (e.g., based on the probability of collisions between mobile broadband (MBB) and low latency transmissions) to facilitate MBB and low latency communication multiplexing with limited signaling.