Abstract: Apparatuses, systems, and techniques to perform computational operations in response to one or more compute uniform device architecture (CUDA) programs. In at least one embodiment, one or more computational operations are to cause one or more other computational operations to wait until matrix multiply-accumulate (MMA) memory transactions are performed.

Abstract: Apparatuses, systems, and techniques to perform computational operations in response to one or more compute uniform device architecture (CUDA) programs. In at least one embodiment, one or more computational operations are to indicate whether matrix multiply-accumulate (MMA) memory operations are complete.

Abstract: Apparatuses, systems, and techniques to perform computational operations in response to one or more compute uniform device architecture (CUDA) programs. In at least one embodiment, one or more computational operations are to cause two or more other computational operations to be performed by two or more streaming multiprocessors (SMs).

Abstract: Apparatuses, systems, and techniques to perform computational operations in response to one or more compute uniform device architecture (CUDA) programs. In at least one embodiment, one or more computational operations are to cause one or more other computational operations to wait until a portion of matrix multiply-accumulate (MMA) operations have been performed.

Abstract: A network of matrix processing units (MPUs) is provided on a device, where each MPU is connected to at least one other MPU in the network, and each MPU is to perform matrix multiplication operations. Computer memory stores tensor data and a master control central processing unit (MCC) is provided on the device to receive an instruction from a host device, where the instruction includes one or more tensor operands based on the tensor data. The MCC invokes a set of operations on one or more of the MPUs based on the instruction, where the set of operations includes operations on the tensor operands. A result is generated from the set of operations, the result embodied as a tensor value.

Abstract: An extruded, directly expanded, high fiber reduced calorie food product, such as a ready-to-eat (RTE) cereal or sweet or savory snack, is produced at high production rates without substantial loss of extrusion functionality and extrudability by replacing a substantial portion of at least one flour with a gelatinized, enzyme-resistant starch type III ingredient or bulking agent as a reduced-calorie, high fiber flour replacer. The resistant starch type III ingredient or bulking agent contains an enzyme-resistant starch type III having a melting point with an endothermic peak temperature of at least about 140° C., and may have a water-holding capacity of less than 3 grams water per gram of the starch-based bulking agent. The total dietary fiber retention of the gelatinized, starch-based bulking agent may be at least about 90% by weight after the extrusion using a die temperature of least about 100° C., and a die pressure of at least about 150 psig.

Abstract: Apparatus for high-volume, high speed ultrasonic cutting or slitting comprising a blade having first and second cutting edges suspended from a reciprocating head, cantilevered over a conveyor supporting and advancing the food product, with the blade including first and second cutting edges on respective opposite sides thereof to cut the food product transversely. The head is preferably controlled for automated operation by an electronic controller to reciprocate transversely at a selected speed and a selected frequency within predetermined ranges over food product advancing on the conveyor. The blade preferably oscillates ultrasonically, and preferably at a frequency of approximately 20,000 Hz. The first and second cutting edges are preferably downwardly convergent, defining an included angle greater than 60 degrees and less than 90 degrees. In one particular embodiment, the internal angle is 70 degrees. In another embodiment, the angle is about 60 degrees.

Abstract: A single-stage process for drying, roasting, and grinding of green coffee beans in a single unit operation in a continuous manner. Heat-treated and ground coffee beans having savory flavor and aroma are achieved in a unique process described herein that combines all these different treatments in a single-stage operation.

Abstract: Apparatus for high-volume, high speed ultrasonic cutting or slitting comprising a blade having first and second cutting edges suspended from a reciprocating head, cantilevered over a conveyor supporting and advancing the food product, with the blade including first and second cutting edges on respective opposite sides thereof to cut the food product transversely. The head is preferably controlled for automated operation by an electronic controller to reciprocate transversely at a selected speed and a selected frequency within predetermined ranges over food product advancing on the conveyor. The blade preferably oscillates ultrasonically, and preferably at a frequency of approximately 20,000 Hz. The first and second cutting edges are preferably downwardly convergent, defining an included angle greater than 60 degrees and less than 90 degrees. In one particular embodiment, the internal angle is 70 degrees. In another embodiment, the angle is about 60 degrees.

Abstract: This invention relates generally to the processing of soy-derived materials for use in various products. More particularly, the invention relates to a process producing highly functional soy protein using ultrafiltration followed by an enzymatic treatment.

Abstract: A method and apparatus for reforming small pieces of natural cheese or other dairy products, such as shreds, trim, or offcuts, wherein the small pieces are fed through a mechanical pump and a die that form the product into a cohesive larger chunk. The mechanical pump and the die operate such that natural cheese pieces may be formed into a larger chunk of natural cheese having a very similar texture and appearance to that of the original product.