Abstract: One embodiment provides for a graphics processing unit to accelerate machine-learning operations, the graphics processing unit comprising a multiprocessor having a single instruction, multiple thread (SIMT) architecture, the multiprocessor to execute at least one single instruction; and a first compute unit included within the multiprocessor, the at least one single instruction to cause the first compute unit to perform a two-dimensional matrix multiply and accumulate operation, wherein to perform the two-dimensional matrix multiply and accumulate operation includes to compute an intermediate product of 16-bit operands and to compute a 32-bit sum based on the intermediate product.

Abstract: Certain aspects of the present disclosure provide techniques for providing an augmented reality user interface, including: receiving, by an image sensor of an electronic device, an image of a physical document; determining a document type associated with the physical document by performing image recognition on the image of the physical document; determining an augmented reality template to display on a display of the electronic device; displaying the augmented reality template on the display of the electronic device, wherein the augmented reality template is aligned in three dimensions with the physical document; determining a distance between the physical document and the electronic device; and enabling one or more interactive user interface elements within the augmented reality template displayed on the display of the electronic device if the determined distance between the physical document and the electronic device is less than a threshold distance.

Abstract: One embodiment provides a graphics processor comprising a memory controller and a graphics processing resource coupled with the memory controller. The graphics processing resource includes circuitry configured to execute an instruction to perform a matrix operation on first input including weight data and second input including input activation data, generate intermediate data based on a result of the matrix operation, quantize the intermediate data to a floating-point format determined based on a statistical distribution of first output data, and output, as second output data, quantized intermediate data in a determined floating-point format.

Abstract: Certain aspects of the present disclosure provide techniques for providing an augmented reality user interface, including: receiving, by an image sensor of an electronic device, an image of a physical document; determining a document type associated with the physical document by performing image recognition on the image of the physical document; determining an augmented reality template to display on a display of the electronic device; displaying the augmented reality template on the display of the electronic device, wherein the augmented reality template is aligned in three dimensions with the physical document; determining a distance between the physical document and the electronic device; and enabling one or more interactive user interface elements within the augmented reality template displayed on the display of the electronic device if the determined distance between the physical document and the electronic device is less than a threshold distance.

Abstract: Projectile loading systems for toy launchers that discharge soft spherical, but tacky, projectiles, the loading systems including a projectile hopper for storing the projectiles, a chute at the bottom of the hopper, the chute having a central groove for lining the projectiles in a single file, an agitator in the hopper. Arrangements of projectile launcher and accessories, container apparatus and the like are provided in differing sizes and orientations with aesthetics of rails in unique angled offset orientations for functions configured with attachment types, bottles, containers, and other apparatus in various combinations.

Abstract: A system can include an elongated member that includes a proximal portion and a distal portion that includes an agitator. The elongated member can extend through a working channel of an endoscope placed within a patient such that the agitator extends past a distal end of the endoscope into a target region within the patient. The agitator can include a plurality of disruption elements that can be in a low-profile state when within the working channel of the endoscope and can transition to an expanded state when advanced past the distal end of the endoscope. The plurality of disruption elements can define an empty cage configuration when in the expanded state. The system can include a driver coupled to the proximal portion of the elongated member. The driver can rotate the elongated member about a longitudinal axis of the elongated member.

Abstract: An apparatus includes a first port set that includes an input port and an output port. The apparatus further includes a plurality of second port sets. Each of the second port sets includes an input port coupled to the output port of the first port set and an output port coupled to the input port of the first port set. The plurality of second port sets are to each communicate at a first maximum bandwidth and the first port set is to communicate at a second maximum bandwidth that is higher than the first maximum bandwidth.

Abstract: Methods, apparatuses, and systems for making prepregs are disclosed. A method may include depositing a resin material onto a surface of a fiber bed and forming a number of discrete resin regions thereon. A distance between the resin regions may be measured to provide desired exposed portions of the surface to facilitate permeation of air through the exposed portions of the surface in a direction perpendicular to a plane of the fiber bed during a curing process of the prepreg.

Abstract: A FPMAC operation has two operands: an input operand and a weight operand. The operands may have a format of FP16, BF16, or INT8. Each operand is split into two portions. The two portions are stored in separate storage units. Then operands are transferred to register files of a PE, with each register file storing bits of an operand sequentially. The PE performs the FPMAC operation based on the operands. The PE may include an FPMAC unit configured to compute an individual partial sum of the PE. The PE may also include an FP adder to accumulate the individual partial sum with other data, such as an output from another PE or an output form another PE array. The FP adder may be fused with the FPMAC unit in a single circuit that can do speculative alignment and has separate critical paths for alignment and normalization.

Abstract: A processing apparatus is provided comprising a multiprocessor having a multithreaded architecture. The multiprocessor can execute at least one single instruction to perform parallel mixed precision matrix operations. In one embodiment the apparatus includes a memory interface and an array of multiprocessors coupled to the memory interface. At least one multiprocessor in the array of multiprocessors is configured to execute a fused multiply-add instruction in parallel across multiple threads.

Abstract: Projectile loading systems for toy launchers that discharge soft spherical, but tacky, projectiles, the loading systems including a projectile hopper for storing the projectiles, a chute at the bottom of the hopper, the chute having a central groove for lining the projectiles in a single file, an agitator in the hopper for disturbing the tacky projectiles in the hopper to separate them, and a projectile transfer structure for carrying a projectile, one at a time, from the chute to a breech or from a feed track during respective priming cycles.

Abstract: An apparatus includes a first port set that includes an input port and an output port. The apparatus further includes a plurality of second port sets. Each of the second port sets includes an input port coupled to the output port of the first port set and an output port coupled to the input port of the first port set. The plurality of second port sets are to each communicate at a first maximum bandwidth and the first port set is to communicate at a second maximum bandwidth that is higher than the first maximum bandwidth.

Abstract: Systems, apparatuses and methods may provide for technology that conduct a first alignment between a plurality of floating-point numbers based on a first subset of exponent bits. The technology may also conduct, at least partially in parallel with the first alignment, a second alignment between the plurality of floating-point numbers based on a second subset of exponent bits, where the first subset of exponent bits are LSBs and the second subset of exponent bits are MSBs. In one example, technology adds the aligned plurality of floating-point numbers to one another. With regard to the second alignment, the technology may also identify individual exponents of a plurality of floating-point numbers, identify a maximum exponent across the individual exponents, and conduct a subtraction of the individual exponents from the maximum exponent, where the subtraction is conducted from MSB to LSB.

Abstract: A system can include a tubular member to be inserted into an esophagus of a patient and a catheter assembly that includes a catheter tube that defines a length greater than a length of the tubular member and is configured to pass through a channel of the tubular member such that a distal tip of the catheter tube extends distally past the distal tip of the tubular member while the tubular member is positioned in the esophagus of the patient. The distal tip of the catheter tube defines a cutting element to core a blockage positioned in the esophagus of the patient. Advancement of the distal tip of the catheter tube into contact with the blockage can core a piece from the blockage that is passed through the catheter tube.

Abstract: A processing apparatus is provided comprising a multiprocessor having a multithreaded architecture. The multiprocessor can execute at least one single instruction to perform parallel mixed precision matrix operations. In one embodiment the apparatus includes a memory interface and an array of multiprocessors coupled to the memory interface. At least one multiprocessor in the array of multiprocessors is configured to execute a fused multiply-add instruction in parallel across multiple threads.

Abstract: Methods, apparatuses, and systems for making prepregs are disclosed. A method may include depositing a resin material onto a surface of a fiber bed and forming a number of discrete resin regions thereon. A distance between the resin regions may be measured to provide desired exposed portions of the surface to facilitate permeation of air through the exposed portions of the surface in a direction perpendicular to a plane of the fiber bed during a curing process of the prepreg.

Abstract: Systems, apparatuses and methods may provide for multi-precision multiply-accumulate (MAC) technology that includes a plurality of arithmetic blocks, wherein the plurality of arithmetic blocks each contain multiple multipliers, and wherein the logic is to combine multipliers one or more of within each arithmetic block or across multiple arithmetic blocks. In one example, one or more intermediate multipliers are of a size that is less than precisions supported by arithmetic blocks containing the one or more intermediate multipliers.

Abstract: One embodiment provides for a graphics processing unit to accelerate machine-learning operations, the graphics processing unit comprising a multiprocessor having a single instruction, multiple thread (SIMT) architecture, the multiprocessor to execute at least one single instruction; and a first compute unit included within the multiprocessor, the at least one single instruction to cause the first compute unit to perform a two-dimensional matrix multiply and accumulate operation, wherein to perform the two-dimensional matrix multiply and accumulate operation includes to compute a 32-bit intermediate product of 16-bit operands and to compute a 32-bit sum based on the 32-bit intermediate product.

Abstract: A system for clearing a blockage in a patient can include a tubular member that defines a channel and is insertable into an esophagus of the patient. The system can include a catheter assembly that includes a catheter tube defining a length greater than a length of the tubular member and being passable through the channel of the tubular member. The catheter tube can include a distal tip that defines a cutting element to core the blockage positioned in the esophagus. The catheter assembly can further include a catheter hub fixedly secured to the catheter tube and couplable with a vacuum line such that, when suction is provided via the vacuum line, advancement of the catheter tube into contact with the blockage cores a piece from the blockage that is passed through the catheter tube.

Abstract: A processing apparatus is provided comprising a multiprocessor having a multithreaded architecture. The multiprocessor can execute at least one single instruction to perform parallel mixed precision matrix operations. In one embodiment the apparatus includes a memory interface and an array of multiprocessors coupled to the memory interface. At least one multiprocessor in the array of multiprocessors is configured to execute a fused multiply-add instruction in parallel across multiple threads.