Abstract: The present invention relates to methods and systems for transmitting and receiving data packets between a first network node and a second network node through a bonded connection. At the first network node, a data packet, a session identification of the data packet and a time value of the data packet are encapsulated in an encapsulating packet. The first network node sends the encapsulating packet from the first network node to the second network node through the bonded connection. The second network node then stores the encapsulating packet after receiving it and determines an expiration time of the encapsulating packet. When the expiration time of the encapsulating packet expired, the second network node dequeues the encapsulating packet and forwards the data packet according to destination of the data packet.

Abstract: An electrochemical apparatus includes an electrolyte, a positive electrode plate, a negative electrode plate, and a separator, where the separator is disposed between the positive electrode plate and the negative electrode plate; a width W of the electrochemical apparatus and a thickness T of the electrochemical apparatus satisfy: 80 mm?W?120 mm, and 2.3?W/T?4; and an adhesion force F between the separator and the positive electrode plate is 5 N/m to 15 N/m. With the adhesion force of the separator being regulated within the foregoing ranges, the separator has high adhesive performance. The application of the separator to a wide and thin electrochemical apparatus with a width and W/T value within the foregoing ranges can mitigate the deformation problem of the wide and thin electrochemical apparatus during cycling while taking the rate performance of the wide and thin electrochemical apparatus into account.

Abstract: A negative active material in the negative active material layer includes hard carbon. An H/C value of the hard carbon is 0.05 to 0.18. A CB value of the electrochemical device is 0.95 to 1.05. This application uses hard carbon as a negative active material. By adjusting the H/C value of the hard carbon and the CB value of the electrochemical device, this application can increase the energy density without lithium precipitation, that is, alleviate lithium precipitation and increase the energy density concurrently.

Abstract: A button cell includes an electrode assembly, a first housing, and a second housing, where the first housing and the second housing are connected and mated to form an accommodating space. The electrode assembly is disposed in the accommodating space. The first housing includes a first bottom wall and a first side wall connected to the first bottom wall. The first bottom wall is provided with a groove. A size of the first bottom wall in a first direction is T ?m, and a size of the groove in the first direction is H ?m, where 0.1T?H?0.7T and the first direction is perpendicular to the first bottom wall.

Abstract: A battery cell includes an electrode assembly and a packaging bag, where the electrode assembly includes a tab, the packaging bag is configured to accommodate the electrode assembly, a sealing zone is formed on a periphery of the packaging bag, one end of the tab is electrically connected to a body of the electrode assembly, and an other end of the tab protrudes out of the sealing zone; the sealing zone is provided with a weak zone, and in a predetermined temperature range, a packaging tension of the weak zone in the packaging bag reduces to release gas inside the battery cell to the outside of the battery cell.

Abstract: The invention provides a detailing process with waterproofing membrane parts and prefabricated accessories which includes below steps: place the first layer membrane on the base template while the edge of the first layer membrane beveled; place the edge of the second layer membrane partially overlapped with the edge of the first layer membrane and then fix the position, and place the membrane edge with bevel to the surface of the second layer membrane fittingly; the edge of the second layer is beveled; and the second bevel is directed towards the direction of the first layer membrane. Press together and make the second chamfered surface fixedly connected to the surface of the first layer. The waterproofing parts connection based on this process has high structural strength, a more complete and unified appearance, improves the waterproofing performance of the parts and accessories, and significantly enhances the service life of the membrane.

Abstract: The present disclosure provides a top cover assembly for a battery, a battery and a device using a battery as a power source. The top cover assembly includes: a cover plate body having a liquid injection hole; a mounting part configured to be connected with the cover plate body and arranged around the liquid injection hole; a sealing element configured to be connected with the mounting part, wherein the sealing element includes a first top wall and a first side wall connected to the periphery of the first top wall, the first top wall is configured to cover the liquid injection hole, and the first side wall is configured to be in press fit with the mounting part to achieve sealing of the liquid injection hole; and a fixing element configured to be connected with the cover plate body and to fix the sealing element to the mounting part.

Abstract: A method for providing input data for a layer of a convolutional neural network (CNN). Input data is received comprising input data values to be processed in a layer of the CNN. Addresses in banked memory of a buffer are determined in which the received data values are to be stored based upon format data indicating a format parameter of the input data in the layer and indicating a format parameter of a filter which is to be used to process the input data in the layer. The received input data values are stored at the determined addresses in the buffer for retrieval for processing in the layer.

Abstract: A haircare appliance having an air inlet, an air outlet, and an airflow generator for generating an airflow from the air inlet to the air outlet and an attachment for a haircare appliance, the attachment including an air inlet, an air outlet. A curved surface adjacent to and downstream of the air outlet. The haircare appliance or the attachment has a curved surface adjacent to and downstream of the air outlet. The haircare appliance or the attachment is configured such that airflow exiting the air outlet generates a first force to attract hair toward the curved surface, and a second force to push hair away from the curved surface.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for training an action selection neural network used to select actions to be performed by an agent interacting with an environment. One of the systems includes (i) a plurality of actor computing units, in which each of the actor computing units is configured to maintain a respective replica of the action selection neural network and to perform a plurality of actor operations, and (ii) one or more learner computing units, in which each of the one or more learner computing units is configured to perform a plurality of learner operations.

Abstract: A vehicle cabin filter assembly comprising a housing and a plurality of filter modules mounted to the housing. The filter modules comprise a first filter module, a second filter module positioned downstream of the first filter module, and a third filter module positioned downstream of the second filter module. Each filter module comprises a particulate filter element, and at least one of the filter modules comprises a gas filter element. The second filter module has a separation efficiency higher than the first filter module, and the third filter module has a separation efficiency higher than the second filter module.

Abstract: A sub-controller (20), controller (30), fluid supply system and apparatus for printing and a method for printing. Provided is a processor controlled sub-controller (20) for controlling the fluid pressure in one or more droplet ejection heads (60); wherein said controller (30) is configured to receive a droplet ejection head movement profile for each of said one or more droplet ejection heads (60), determine a respective induced fluid pressure profile at one or more predetermined locations for each of said one or more droplet ejection heads (60) using the respective droplet ejection head movement profile; and generate respective pressure correction data for each of said one or more droplet ejection heads based on the respective induced fluid pressure profile and a predetermined pressure window to be maintained at said one or more droplet ejection heads (60).

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for implementing associative memory. In one aspect a system comprises an associative memory neural network to process an input to generate an output that defines an energy corresponding to the input. A reading subsystem retrieves stored information from the associative memory neural network. The reading subsystem performs operations including receiving a given, i.e. query, input and retrieving a data element from the associative memory neural network that is associated with the given input. The retrieving is performed by iteratively adjusting the given input using the associative memory neural network.

Abstract: The present disclosure proposes the use of a dual discriminator network that comprises a temporal discriminator network for discriminating based on temporal features of a series of images and a spatial discriminator network for discriminating based on spatial features of individual images. The training methods described herein provide improvements in computational efficiency. This is achieved by applying the spatial discriminator network to a set of one or more images that have reduced temporal resolution and applying the temporal discriminator network to a set of images that have reduced spatial resolution. This allows each of the discriminator networks to be applied more efficiently in order to produce a discriminator score for use in training the generator, whilst maintaining accuracy of the discriminator network. In addition, this allows a generator network to be trained to more accurately generate sequences of images, through the use of the improved discriminator.

Abstract: Shader processing units for a graphics processing unit that are configured to execute one or more ray tracing shaders that generate ray data associated with one or more rays. The ray data for a ray includes a plurality of ray data elements. The shader processing unit comprises local storage, and store logic. The store logic is configured to receive, as part of a ray tracing shader, a ray store instruction that comprises: (i) information identifying a store group of a plurality of store groups, each store group of the plurality of store groups comprising one or more ray data elements of the plurality of ray data elements, and (ii) information identifying one or more ray data elements of the identified store group to be stored in an external unit). In response to receiving the ray store instruction, the store logic retrieves the identified ray data elements for one or more rays from the storage.

Abstract: A circuit for estimating a time difference between a first signal and a second signal includes a first signal line for receiving the first signal; a delay unit configured to receive the second signal and delay the second signal so as to provide a plurality of delayed versions of the second signal, each delayed version being delayed by a different amount of delay to the other delayed versions; a comparison unit configured to compare each of the delayed versions of the second signal with the first signal so as to identify which of the delayed versions of the second signal is the closest temporally matching signal to the first signal; and a difference estimator configured to estimate the time difference between the first and second signals in dependence on the identified delayed version.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for selecting action slates using reinforcement learning. One of the methods includes receiving an observation characterizing a current state of an environment; selecting an action slate by processing the observation and a plurality of candidate action slates using a deep neural network, wherein each candidate action slate comprises a respective plurality of actions from the set of actions, and wherein the deep neural network is configured to, for each of the candidate action slates, process the observation and the actions in the candidate action slate to generate a slate Q value for the candidate action slate that is an estimate of a long-term reward resulting from the candidate action slate being provided to the action selector in response to the observation; and providing the selected action slate to an action selector in response to the observation.

Abstract: An electrode assembly is disposed within the housing and includes a first electrode plate, a second electrode plate, and a separator disposed between the first electrode plate and the second electrode plate. Along the first direction, a projection of the second electrode plate is within a projection of the first electrode plate, the projection of the first electrode plate is within a projection of the separator, and the projection of the separator is within a projection of the accommodating space. Along a direction perpendicular to the first direction, the separator extends beyond the first electrode plate, and an end of a portion of the separator extending beyond the first electrode plate, together with the housing enclose a first region, a width of the first region is W1, wherein W1>0.

Abstract: An electrochemical apparatus includes an electrolyte, a positive electrode plate, a negative electrode plate, and a separator. The separator is disposed between the positive electrode plate and the negative electrode plate. The separator includes a substrate and a first coating provided on both surfaces of the substrate. The first coating includes polymer particles. In an 11.5 ?m×7.5 ?m region of a surface of the first coating, a quantity Q of the polymer particles is 10 to 90. The electrolyte includes substituted or unsubstituted C3-C8 linear carboxylate, and in a case of substitution, a substituent group is selected from halogen atoms. A mass percentage B of the linear carboxylate is 5% to 50% based on a mass of the electrolyte.