Abstract: Techniques are disclosed that enable training a plurality of policy networks, each policy network corresponding to a disparate robotic training task, using a mobile robot in a real world workspace. Various implementations include selecting a training task based on comparing a pose of the mobile robot to at least one parameter of a real world training workspace. For example, the training task can be selected based on the position of a landmark, within the workspace, relative to the pose. For instance, the training task can be selected such that the selected training task moves the mobile robot towards the landmark.

Abstract: An analysis method for the causal inference of human physiological network in multiscale time series signals includes the following steps: S1: decomposing physiological signals u1, u2, . . . , um to be analyzed by using a noise-assisted multivariate empirical mode decomposition (NA-MEND) algorithm; S2: carrying out a causal analysis between two different physiological signals ui, uj, where i=1, 2, . . . , m, j=1, 2, . . . , m, and i?j, to obtain a causality between the two signals; and S3: repeating step S2 for any two signals in u1, u2, . . . , um until a causality between each two signals in u1, u2, . . . , um is obtained to form the causal network. The present invention can effectively analyze the causal network of the physiological signals, thereby facilitating the application of the physiological signals.

Abstract: A semiconductor structure is provided which includes a nanosheet stack structure on a base. The nanosheet stack structure includes a multilayered nanosheet between adjacent nanosheet layers. The multilayered nanosheet includes one or more first layers of a first material and one or more second layers of a second material, wherein the first material has an etch selectivity different than the second material. The one or more first layers of the multilayered nanosheet are recessed. A first inner spacer includes a third material is formed by depositing the third material into an outer portion of the one or more recessed first layers of the multilayered nanosheet. The one or more second layers of the multilayered nanosheet are recessed. A second inner spacer includes a fourth material which is formed by depositing the fourth material into an outer portion of the one or more recessed second layers of the first multilayered nanosheet.

Abstract: A punching abnormality detection system, comprising: a punch, generating a vibration signal when working; an ultrasonic sensor, fixed on the punch and configured to convert the vibration signal into a voltage response signal; a data acquisition device, electrically connected to the ultrasonic sensor, configured to collect the voltage response signal in real time, generate a monitoring envelope curve, and analyze whether the voltage response curve deviates from the monitoring envelope curve; when the voltage response curve exceeds the monitoring envelope curve, the data acquisition device determine that the punch works abnormally, and an abnormal signal is output, thereby predicting production interruption caused by equipment abnormality during the production process and improving production quality.

Abstract: An ultrasonic sensor, which is installed on a test device, includes a shell, one end is open, the other end is closed surface; a cover sheet, bonded to the open end of the shell; a copper foil, a lower surface of the copper foil is bonded to an inner bottom surface of the closed surface of the shell; a piezoelectric chip, a lower surface of the piezoelectric chip is bonded to a upper surface of the copper foil, and when the test device vibrates, the piezoelectric chip converts a vibration signal into a voltage response signal; a cable, a positive electrode is welded on a upper surface of the piezoelectric chip, and a negative electrode is welded on the upper surface of the copper foil, configured to connect to an external detection device and transmit the voltage response signal to the external detection device.

Abstract: A computer system and method for answering a natural language question is provided. The system comprises at least one processor and a memory storing instructions which when executed by the processor configure the processor to perform the method. The method comprises receiving a natural language question, generating a SQL query based on the natural language question, generating an explanation regarding a solution to the natural language question as answered by the SQL query, and presenting the solution and the explanation.

Abstract: A method of forming stacked fin field effect devices is provided. The method includes forming a layer stack on a substrate, wherein the layer stack includes a first semiconductor layer on a surface of the substrate, a second semiconductor layer on the first semiconductor layer, a third semiconductor layer on the second semiconductor layer, a separation layer on the third semiconductor layer, a fourth semiconductor layer on the separation layer, a fifth semiconductor layer on the fourth semiconductor layer, and a sixth semiconductor layer on the fifth semiconductor layer. The method further includes forming a plurality of channels through the layer stack to the surface of the substrate, and removing portions of the second semiconductor layer and fifth semiconductor layer to form lateral grooves.

Abstract: An example method can comprise creating a non-linear neural network based model of a system based on historical operational data of the system and receiving first sensor data from a plurality of sensors associated with the system. Predicted next sensor data can be determined based on the received first sensor data and the non-linear network model. Second sensor data can be received from the plurality of sensors, and a measure of deviation between the predicted next sensor data and the received second sensor data is calculated. In response to the measured deviation exceeding a predefined threshold; it can be determined that a fault has occurred.

Abstract: A fully automatic online aero aluminum sorting and recovery system based on LIBS (Laser Induced Breakdown Spectroscopy) technology, which belongs to the field of aero aluminum sorting and recovery technology, and is suitable for online sorting, detection and recovery of large batch of aero aluminum. The fully automatic online aero aluminum sorting system based on LIBS technology provided in the present invention consists of six portions: a sample feeding unit (1), a surface treatment unit (2), a material positioning unit (3), a LIBS analysis and detection unit (4), a transfer unit (5) and a sorting and recovery unit (6). The system according to the invention can be used to realize the automatic online detection, sorting and recovery of aero aluminum, and the system does not have requirements on the surface condition of the recovered aero aluminum samples. The sorting accuracy rate is greater than 90% and the sorting rate is not less than 1 block per second.

Abstract: A method for manufacturing a semiconductor device includes forming a first semiconductor layer on a semiconductor substrate, forming a second semiconductor layer including a first concentration of germanium on the first semiconductor layer, and forming a third semiconductor layer on the second semiconductor layer. The first and third semiconductor layers each have a concentration of germanium, which is greater than the first concentration of germanium. The first, second and third semiconductor layers are patterned into at least one fin. The method further includes covering the second semiconductor layer with a mask layer. In the method, a bottom source/drain region and a top source/drain region are simultaneously grown from the first semiconductor layer and the third semiconductor layer, respectively. The mask layer is removed from the second semiconductor layer, and a gate structure is formed on and around the second semiconductor layer.

Abstract: The present invention relates to a method and a system for robust, searchable, symmetrical encryption, the method comprising: during initialization, having a retrieval server perform retrieval on stored ciphertext information according to a searchable ciphertext and retrieval trapdoor information (Cntupd, K, Locgrp) sent by a client, wherein, the retrieval server acquires a last received first ciphertext set Cnew from a ciphertext database CDB just before the present retrieval according to a retrieval key K and an encryption counter Cntupd sent by the client, acquiring a second ciphertext set Ccache in a retrieval request history from a cache ciphertext database GRP according to a cache ciphertext index Locgrp, and merging the first ciphertext set Cnew and the second ciphertext set Ccache to form a retrieval set , and sending it to the client for decryption.

Abstract: The disclosure discloses a touch display panel and a touch display device. The touch display panel includes a cathode and a number of touch electrodes, wherein an interior of at least one touch electrode is hollowed out to reduce a relative area between the cathode and the touch electrode. The relative area between the touch electrode and the cathode of the touch display panel may be reduced by hollowing out the interior of at least one touch electrode, thereby reducing the inductive capacitance between the touch electrode and the cathode.

Abstract: An optical gateway includes a light sensor unit, comprising at least one light sensor, configured to sense state information of a signal lamp of an external test device; a control unit, connected to the light sensor unit, and configured to obtain the state information of the signal lamp; and a communication unit, connected to the control unit and configured to send the status information of the signal lamp to an external cloud server. The invention also provides a remote monitoring system, and the user can obtain the state information of the external test device from the cloud server, and monitor the working state of the external test device in real time through the optical gateway and the remote monitoring system.

Abstract: An improvement in a nuclear fuel rod is disclosed. The improved fuel rod includes a cladding tube, a plurality of fuel pellets stacked within the cladding tube, and a liquid material filling the gap between the fuel pellets and the cladding tube. The liquid material is selected from those having a thermal conductivity higher than that of helium, a melting point lower than about 400° C., a boiling point higher than 1600° C., and which are capable of wetting both the fuel pellets and the cladding sufficient to form a protective layer over the pellets and to wick into openings that may form in the cladding.

Abstract: The present disclosure provide a method for identifying power equipment targets based on human-level concept learning, including: creating a dataset of power equipment images, and annotating power equipment in power equipment images; training neural network and Bayesian network with the annotated dataset and respectively acquire identification results and conditional probabilities; calculating probabilities of unions with the conditional probabilities; and filtering the identification result corresponding to the highest probability of the union as identification result of the dataset of the power equipment images and complete the identification of the power equipment. The present disclosure combines Mask R-CNN and probabilistic graphical model. The bottom layer uses Mask R-CNN, and the top layer uses Bayesian network to train in identifying power equipment images, so that a small amount of data samples can achieve good recognition, which improved the performance of Mask R-CNN model.

Abstract: Semiconductor devices and methods of forming a first layer cap at ends of layers of first channel material in a stack of alternating layers of first channel material and second channel material. A second layer cap is formed at ends of the layers of second channel material. The first layer caps are etched away in a first device region. The second layer caps are etched away in a second device region.

Abstract: A method of forming a water resistant boundary on a fissile material for use in a water cooled nuclear reactor is described. The method comprises mixing a powdered fissile material selected from the group consisting of UN and U3Si2 with an additive selected from oxidation resistant materials having a melting or softening point lower than the sintering temperature of the fissile material, pressing the mixed fissile and additive materials into a pellet, sintering the pellet to a temperature greater than the melting point of the additive.

Abstract: A semiconductor device includes one or more fins. Each fin includes a top channel portion formed from a channel material, a middle portion, and a bottom substrate portion formed from a same material as an underlying substrate. An oxide layer is formed between the bottom substrate portion of each fin and the isolation layer, with a space between a sidewall of at least a top portion of the isolation dielectric layer and an adjacent sidewall of the one or more fins, above the oxide layer. A gate dielectric, protruding into the space and in contact with the middle portion, is formed over the one or more fins and has a portion formed from a material different from the oxide layer.

Abstract: An oral care composition, including a mixture of antibacterial agents, the mixture of antibacterial agents including oleanic acid and eugenol at a 2:1 to 3:1 weight ratio.

Abstract: A media management operation to write data from a source block of a cache memory to a set of pages of a destination block of a storage area of a memory sub-system that is at a higher data density than the cache memory a write request to program data to a memory device of a memory sub-system is executed. An entry of a first data structure identifying a page count corresponding to the source block of the cache memory is generated. Following a determination that the data is written to the set of pages of the destination block of the storage area, the entry is updated to identify a decreased page count corresponding to the source block, where the data is erased from the source block when the decreased page count satisfies a condition. A second entry of a second data structure including information mapping a logical block to the source block of the cache memory is also updated.