Abstract: A method, a device and electronic device for clarification question generation are provided in one or more embodiments of this disclosure. The method includes: extracting entity information from a fuzzy context input by a user; inputting the fuzzy context into a template generating module of a pre-built CQG neural network model so as to obtain a clarification question template; inputting the entity information into an entity rendering module of the CQG neural network model so as to obtain at least one entity phrase; and generating a clarification question for a fuzzy question based on the clarification question template and the at least one entity phrase for presenting to the user.

Abstract: A method and system for constructing a lightweight container-based user environment (CUE), and a medium, the method including: preparing, by a main process, for communication, cloning a child process, and then becoming a parent process; elevating, by the child process, permission, executing namespace isolation, and cloning a grandchild process, and setting, by the parent process, cgroups for the grandchild process; and setting, by the grandchild process, permission of the grandchild process to execute a command and a file, preparing an overlay file system, setting a hostname, restricting permission, and executing an initialization script to start the container. Multiple users are allowed to customize their own environments, enabling the users to customize their environments more flexibly, achieving privacy isolation, and making it easier and more secure to update a system. Therefore, it is particularly applicable to a high-performance computing cluster.

Abstract: The invention discloses a device and text representation method applied to sentence embedding, which has determining a parent word and a child words set corresponding to the parent word, obtaining an hidden interaction state of parent word based on the hidden interaction state of all child words; obtaining parent word sequence corresponding to the parent word and obtaining hidden state sequence corresponding to the hidden state of parent word; obtaining the interaction representation sequence of each parent word and other parent word based on the hidden state sequence, generating sentence embeddings. The invention proposes to realize sentence embeddings through a two-level interaction representation.

Abstract: A battery energy storage system (BESS) including: a battery storage configured to store surplus renewable energy; a controller coupled to the battery storage and configured to control charging operations and discharging operations of the battery storage; a generation meter coupled to the controller and configured to measure renewable energy; a renewable energy generator coupled to the generation meter and configured to generate renewable energy; and a standard meter coupled to the controller and configured to measure energy provided by a power grid; wherein the controller is configured to manage energy expenditure of the BESS according to the following steps: initializing a replay buffer configured to store state transition samples; initializing a main net configured to generate a current Q-value; initializing a target net configured to generate a target Q-value; obtaining an environment state of the BESS; and selecting an action based on an ?-greedy policy.

Abstract: A system for streaming data items of a streaming application via a Vertical Cuckoo Filter (VCF), including: one or more computers communicating with a provider via a network; wherein the one or more computers includes: a processor; and a storage device coupled to the processor, wherein the storage device is configured to store the VCF, an insert module, a lookup module, and a delete module, the VCF includes buckets with one or more slots and includes at least two bitmasks, the insert module is configured to perform an insert operation of a first fingerprint of a first item from the provider on the VCF, the lookup module is configured to perform a lookup operation of a second fingerprint of a second item from the provider on the VCF, the delete module is configured to perform a delete operation of a third fingerprint of a third item from the VCF.

Abstract: A Mobile-Assisted edge computing framework including: one or more requests to be processed; an operator configured to assign the one or more requests to a static edge node; and a cloud configured to cache and pre-fetch services for the one or more requests; wherein the one or more requests include handled requests and unhandled requests, the static edge node is configured to process the handled requests, and one or more mobile edge nodes are configured to process the unhandled requests; and wherein the static edge node is configured to auction the unhandled requests to the one or more mobile edge nodes, the auction including: generating a candidate set of the one or more mobile edge nodes; assigning an unhandled request to a target mobile edge node of the candidate set; and processing the unhandled request via the target mobile edge node.

Abstract: An intelligent control gas suction-type electric propulsion system applicable to multi-flow regimes: an ultra-low orbit rare gas is used as a working medium for attitude orbit control and resistance compensation propulsion, the gas is collected and inputted into an intelligent feedback pressurization system by means of a parabolic gas intake duct, intelligent feedback and pressurization are performed on the gas working medium by a molecular pump and a gas pump and the medium is stored in a working fluid storage tank so as to supply a hybrid thruster system that consists of seven sets of electric thrusters to generate thrust, which may achieve multiple thrust modes, and achieve the purpose of attitude orbit control and resistance compensation.

Abstract: An inductive plasma acceleration apparatus, comprising a pulse laser assembly, a pulsed discharge assembly, an exciting coil assembly, a solid state working medium, and a control assembly; the exciting coil assembly is electrically connected to the pulsed discharge assembly such that a strong pulse current is produced in the exciting coil assembly during the discharge process of the pulse discharge assembly, and an inductive pulse electromagnetic field is excited around the exciting coil assembly; the solid state working medium is positioned on the optical path of a pulse laser emitted by the pulse laser assembly such that the solid state working medium produces a pulse gas under the ablation action of the pulse laser, and the inductive pulse electromagnetic field is positioned on the circulation gas path of the pulse gas such that the pulse gas can enter the inductive pulse electromagnetic field.

Abstract: A high-performance computing-oriented method for automatically deploying an execution environment along with a job, including: presetting isolated execution environments at nodes of a high-performance computing system; logging in an isolated execution environment of a login node; carrying out development and debugging on the job and configuration on a job execution environment at the login node, and issuing a job running request to a job management system; assigning compute nodes from the nodes of the high-performance computing system to the job of the user by the job management system, automatically deploying an file system of the user synchronously to the assigned compute nodes along with the job when the job is loaded, and running the job of the user by the corresponding compute nodes; and feeding results back to the login node of the user after running the job is completed, then clearing file systems.

Abstract: This invention discloses a hierarchical switching fabric and deadlock avoidance method for ultra high radix network routers. The hierarchical switching fabric comprises a network-on-chip and K multi-port components. The multi-port component comprises a port module configured to receive packets by a high speed serializer/deserializer, code and format the packets, send the packets to a corresponding hyper packet module after coding and format conversion, and send the packets sent by the hyper packet module to the network; and the hyper packet module configured to perform protocol processing for the received data link level packets, discard illegal packets, forward legitimate packets to the network-on-chip, perform data error correcting, format conversion and channel mapping for the packets received from the network-on-chip, and send the packets to the corresponding port module.

Abstract: The present disclosure provides a visual localization method and apparatus based on a semantic error image.

Abstract: A flexible variable emissivity electrochromic device and a preparation method thereof are disclosed. The device includes a working electrode, a gel electrolyte layer, and a counter electrode sequentially from top to bottom. The working electrode includes a flexible polymer film and a metal film, the flexible polymer film is a surface-modified film and/or a film with a transition layer plated on a lower side thereof, and the metal film is deposited on the surface-modified film or the transition layer. The electrolyte layer includes a porous membrane and an electrolyte. The electrolyte is infiltrated in the porous membrane. The electrolyte includes an electrochromic material containing metal ions and a solvent, the metal ions enable reversible electrodeposition and dissolution, and metal of the metal ions is different from that used in the metal film. The preparation method includes preparing and assembling a working electrode, a gel electrolyte layer and a counter electrode.

Abstract: This invention discloses a multimode interconnection interface controller for a converged network, which comprises a SERDES element responsible for serial/parallel conversion, a LANE_TXCLK element responsible for generating a transmit clock, a SERDES initialization element responsible for link training and rate negotiation, and a PCS_EB coding element and an PCS_AF coding element responsible for physical layer coding of messages. The link training and rate negotiation are completed automatically via the shared SERDES initialization element. More universality and flexibility are provided for interconnection chip design by the PCS_EB coding element. The PCS_AF coding element is provided to reduce message penetration delay. The multimode interconnection interface controller is integrated in a single chip. Through flexible configuration, the single chip can meet transmission requirements of dedicated high speed networks and Ethernet networks.

Abstract: The present disclosure provides an item recommendation method and system based on a user intention in a conversation session. In this method, a current conversation session is encoded by use of items of prior time sequence and the last item so as to generate an accurate conversation session representation which is then used to perform initial item prediction as a user intention. Afterwards, an intention-guided neighbor detector is used to locate an accurate neighbor conversation session. Finally, the current conversation session and the neighbor conversation session representations are combined adaptively by a gated mixed layer to generate the final item recommendation, so as to effectively improve the accuracy of item recommendation.

Abstract: A wafer-level assembly method for a micro hemispherical resonator gyroscope includes: after independently manufactured glass substrates are softened and deformed at a high temperature, forming a micro hemispherical resonator on the glass substrate; forming glass substrate alignment holes at both ends of the glass substrate by laser ablation; aligning and fixing a plurality of identical micro hemispherical resonators on a wafer fixture by using the alignment holes as a reference, and then performing operations by using the wafer fixture as a unit to implement subsequent processes that include: releasing the micro hemispherical resonators, metallizing the surface, fixing to the planar electrode substrates, separating the wafer fixture and cleaning to obtain a micro hemispherical resonator gyroscope driven by a bottom planar electrode substrate.

Abstract: The present disclosure relates to an online task dispatching and scheduling system. The system includes an end device; an access point (AP) configured to receive a task from the end device; one or more edge servers configured to receive the task from the AP, the one or more edge servers including a task waiting queue, a processing pool, a task completion queue, and a scheduler, wherein the AP further includes a dispatcher utilizing Online Learning (OL) for determining a real-time state of network conditions and server loads; and the AP selects a target edge server from the one or more edge servers to which the task is to be dispatched; and wherein the scheduler utilizes Deep Reinforcement Learning (DRL) in generating a task scheduling policy for the one or more edge servers.

Abstract: The present disclosure relates to an edge federation system, wherein the edge federation system includes: an end device; an edge service provider (ESP) configured to contract services from the end device; an edge infrastructure provider (EIP) configured to provide edge computing resources to the contracted services from the ESP; a cloud server configured to provide service to the end device; and a consortium for managing resources between the EIP and the cloud server, wherein the consortium is further configured to calculate available resources of the EIP, calculate available resources of the cloud server, and optimize the distribution of services from the EIP and the cloud server to the end device.

Abstract: The invention provides a converged network interface card, a message coding method and a message transmission method thereof. The converged network interface card comprises a PCIE host interface processing module, a high speed network card core logic, a crossbar switch XBAR, an Ethernet network card core logic, an Ethernet message dicing/slicing module, a physical layer, a high speed network/Ethernet message conversion module EoH, and a high speed network/Ethernet configurable network port. The invention supports customized high speed interconnection interface and a standard Ethernet interface on a set of network hardware, and supports three working modes on a set of physical hardware (high speed network mode, Ethernet mode and EoH mode transmitting Ethernet messages over the high speed network), implements seamless compatibility between the high speed network/Ethernet, and flexibly supports multimode applications such as scientific computing and cloud computing.

Abstract: This invention discloses a multimode interconnection interface controller for a converged network, which comprises a SERDES element responsible for serial/parallel conversion, a LANE_TXCLK element responsible for generating a transmit clock, a SERDES initialization element responsible for link training and rate negotiation, and a PCS_EB coding element and an PCS_AF coding element responsible for physical layer coding of messages. The link training and rate negotiation are completed automatically via the shared SERDES initialization element. More universality and flexibility are provided for interconnection chip design by the PCS_EB coding element. The PCS_AF coding element is provided to reduce message penetration delay. The multimode interconnection interface controller is integrated in a single chip. Through flexible configuration, the single chip can meet transmission requirements of dedicated high speed networks and Ethernet networks.

Abstract: This invention discloses a hierarchical switching fabric and deadlock avoidance method for ultra high radix network routers. The hierarchical switching fabric comprises a network-on-chip and K multi-port components. The multi-port component comprises a port module configured to receive packets by a high speed serializer/deserializer, code and format the packets, send the packets to a corresponding hyper packet module after coding and format conversion, and send the packets sent by the hyper packet module to the network; and the hyper packet module configured to perform protocol processing for the received data link level packets, discard illegal packets, forward legitimate packets to the network-on-chip, perform data error correcting, format conversion and channel mapping for the packets received from the network-on-chip, and send the packets to the corresponding port module.