Abstract: A control method for a reinforcement learning system includes following operations. The reinforcement learning system obtains training data relating to an interaction system. The interaction system interacts with a reinforcement learning agent. A neural network model is utilized by the reinforcement learning agent for selecting sequential actions from a set of candidate actions. The neural network model is trained to maximize cumulative rewards collected by the reinforcement learning agent in response to the sequential actions. During training of the neural network model, auxiliary rewards of the cumulative rewards are provided to the reinforcement learning agent according to a comparison between symptom inquiry actions of the sequential actions and diagnosed symptoms in the training data.

Abstract: A protective sleeve designed to prevent the passage of bodily fluids therethrough. The protective sleeve includes an elongated member having an interior and an exterior. The interior of the elongated member has a perimeter, wherein the perimeter defines a top side, a bottom side, and at least one side edge. The top side and the bottom side each have an elastic band disposed therein. The elongated member is designed to seal around an arm of a user and thereby prevent fluid penetration therethrough. In this way, a user is able to protect their arm from exposure to bodily fluids, such as blood.

Abstract: Systems, apparatuses, and methods for efficiently performing operations system are disclosed. A computing system uses a memory for storing data, and one or more processing units. The memory includes multiple rows for storing the data with each intersection of a row and a column being a memory bit cell. The memory processes operations. For particular operations, the two or more operands are accessed simultaneously for generating a result without being read out and stored. Two indications are generated specifying at least a first row and a second row targeted by the operation. The memory generates a result by performing the operation for each of the one or more cells in the first row a stored value with a respective stored value in the one or more cells in the second row.

Abstract: An apparatus is described. The apparatus includes an image processing unit. The image processing unit includes a plurality of stencil processor circuits each comprising an array of execution unit lanes coupled to a two-dimensional shift register array structure to simultaneously process multiple overlapping stencils through execution of program code. The image processing unit includes a plurality of sheet generators respectively coupled between the plurality of stencil processors and the network. The sheet generators are to parse input line groups of image data into input sheets of image data for processing by the stencil processors, and, to form output line groups of image data from output sheets of image data received from the stencil processors. The image processing unit includes a plurality of line buffer units coupled to the network to pass line groups in a direction from producing stencil processors to consuming stencil processors to implement an overall program flow.

Abstract: An image processing method includes generating, by a processing component, a first input feature map based on an input image using a first convolutional neural network; generating, by the processing component, a first template feature map based on a template image using the first convolutional neural network; generating, by the processing component, a first estimated motion parameter based on an initial motion parameter, the first input feature map and the first template feature map using an iterative Lucas-Kanade network; and performing, by the processing component, image alignment between the input image and the template image based on the first estimated motion parameter.

Abstract: A method for adjusting a convolutional neural network includes following operations. The convolutional neural network includes convolution layers in a sequential order. Receptive field widths of the convolution layers in a first model of the convolutional neural network are determined. Channel widths of the convolution layers in the first model are reduced into reduced channel widths according to the receptive field widths of the convolution layers and an input image width. A structure of a second model of the convolutional neural network is formed according to the reduced channel widths. The second model of the convolutional neural network is trained according the structure of the second model.

Abstract: A computing device for training a fully-connected neural network (FCNN) comprises at least one storage device; and at least one processing circuit, coupled to the at least one storage device. The at least one storage device stores, and the at least one processing circuit is configured to execute instructions of: computing a block-diagonal approximation of a positive-curvature Hessian (BDA-PCH) matrix of the FCNN; and computing at least one update direction of the BDA-PCH matrix according to an expectation approximation conjugated gradient (EA-CG) method.

Abstract: An electronic apparatus includes a displayer, a graphic processing circuit, sensors and a control circuit. The displayer is configured for displaying a virtual reality scene or an augmented reality scene. The graphic processing circuit is coupled to the displayer. The graphic processing circuit is configured for computing a plurality of scene segments in the virtual reality scene or the augmented reality scene. The sensors are configured for collecting attention factors. The control circuit is coupled to the sensors and the graphic processing circuit. The control circuit is adapted to generate an attention map according to the attention factors. The attention map indicates priority rankings of the scene segments. The control circuit is further adapted to allocate a distribution of computational resources of the graphic processing circuit between the scene segments according to the priority rankings. The scene segments are computed differentially according to the distribution of computational resources.

Abstract: A machine learning system includes a memory and a processor. The processor is configured to access and execute at least one instruction from the memory to perform inputting raw data to a first partition of a neural network, in which the first partition at least comprises an activation function of the neural network. The activation function is applied to convert the raw data into irreversible metadata. The metadata is transmitted to a second partition of the neural network as inputs to generate a learning result corresponding to the raw data.

Abstract: A system and method for attitude correction is provided. An acceleration and an attitude of an electronic device are detected. A period of time where a velocity of the electronic device at the beginning of the period of time and a velocity of the electronic device at the end of the period of time are equal is identified. An attitude correction is calculated based on the identified period of time and the detected acceleration of the electronic device during the period of time. The detected attitude of the electronic device is corrected with the calculated attitude correction.

Abstract: A cap may include a top and a sidewall depending from the top. A plug may also extend from the top, spaced from the sidewall. Threads are formed either on an internal surface of the sidewall, i.e., generally facing the plug, or on an external surface of the plug, i.e., generally facing the sidewall. The spacing between the sidewall and the plug generally defines a receptacle configured for receiving an open end of a container. The sidewall and the plug may be angled relative to each other, such that the receptacle has a cross-section that is generally V-shaped.

Abstract: The present disclosure provides a method, an electronic device, and a user interface for searching a plurality of relevant via active searching. At first, a query image having a first subject and a second subject may be obtained from stored images. Next, a query concept based on content information of the query image may be learned by sampling first sample images from the stored images according to content information of the query image for relevance feedback. Then, the query concept is refined based on context information of the first selected images among the first sample images. Based on the refined query concept, relevant images among the stored images may be searched according to the query concept and grouped into a collection album.

Abstract: A computer aided medical method includes the following steps. An initial symptom of a patient and context information is collected through an interaction interface. Actions in a series are sequentially generated according to the candidate prediction models and the initial symptom. Each of the actions corresponds to one of the inquiry actions or one of the disease prediction actions. If the latest one of the sequential actions corresponds to one of the disease prediction actions, potential disease predictions are generated in a first ranking evaluated by the candidate prediction models. The first ranking is adjusted into a second ranking according to the context information. A result prediction corresponding to the potential disease predictions is generated in the second ranking.

Abstract: A computer aided medical method include following steps. An initial symptom is collected through an interaction interface. A representative prediction model is selected from plural candidate prediction models according to the initial symptom. The candidate prediction models are trained by a machine learning algorithm according to clinical data. A series of sequential actions is generated according to the representative prediction model and the initial symptom. The sequential actions are selected from plural candidate actions in the representative prediction model. The candidate actions include plural inquiry actions and plural disease prediction actions. Each of the sequential actions is one of the inquiry actions or the disease prediction actions. The series of sequential actions is displayed on the interaction interface.

Abstract: A training task optimization system includes a processor. The processor is configured to receive training environment information of a training task. The training environment information at least carries information corresponding to training samples in the training task. The processor is configured to calculate a memory distribution for the training task based on memory factors, the training samples and a neural network, and select a mini-batch size that is fit to the memory distribution. In response to the training environment information, the processor is configured to output the mini-batch size for execution of the training task.

Abstract: An image processor unit is described. The image processor unit includes a plurality of inputs to receive at least one input image. The image processor unit includes a plurality of outputs to provide at least one output image. The image processor unit includes a network coupled to the plurality of inputs and the plurality of outputs. The network is to couple at least one of the inputs to at least one of the outputs. The image processor unit includes an image processor circuit coupled to the network. The network to route an input image that is received at one of the inputs to the image processor circuit. The image processor circuit is to execute image signal processing program code to generate a processed output image from the input image. The network is to route the processed output image to at least one of the outputs.

Abstract: An image processing training method includes the following steps. A template label image is obtained, in which the template label image comprises a label corresponding to a target. A plurality of first reference images are obtained, in which each of the first reference images comprises object image data corresponding to the target. A target image according to the template label image and the first reference images is generated, in which the target image comprises a generated object, a contour of the generated object is generated according to the template label image, and a color or a texture of the target image is generated according to the first reference images.

Abstract: An electronic apparatus includes a displayer, a graphic processing circuit, sensors and a control circuit. The displayer is configured for displaying a virtual reality scene or an augmented reality scene. The graphic processing circuit is coupled to the displayer. The graphic processing circuit is configured for computing a plurality of scene segments in the virtual reality scene or the augmented reality scene. The sensors are configured for collecting attention factors. The control circuit is coupled to the sensors and the graphic processing circuit. The control circuit is adapted to generate an attention map according to the attention factors. The attention map indicates priority rankings of the scene segments. The control circuit is further adapted to allocate a distribution of computational resources of the graphic processing circuit between the scene segments according to the priority rankings. The scene segments are computed differentially according to the distribution of computational resources.

Abstract: A cap may include a top and a sidewall depending from the top. A plug may also extend from the top, spaced from the sidewall. Threads are formed either on an internal surface of the sidewall, i.e., generally facing the plug, or on an external surface of the plug, i.e., generally facing the sidewall. The spacing between the sidewall and the plug generally defines a receptacle configured for receiving an open end of a container. The sidewall and the plug may be angled relative to each other, such that the receptacle has a cross-section that is generally V-shaped.