Abstract: Apparatuses, systems, and techniques to cause one or more portions of one or more neural networks to be trained. In at least one embodiment, one or more portions of one or more neural networks are caused to be trained by, for example, iteratively adjusting precision of weight parameters associated with the one or more portions based, at least in part, on one or more performance metrics of the one or more portions.

Abstract: Apparatuses, systems, and techniques to identify bounding boxes of objects with in an image. In at least one embodiment, bounding boxes are determined in an image using an intersection over union threshold that is based at least in part on a size of an object.

Abstract: Apparatuses, systems, and techniques to perform non-maximum suppression (NMS) in parallel to remove redundant bounding boxes. In at least one embodiment, two or more parallel circuits to perform two or more portions of a NMS algorithm in parallel to remove one or more redundant bounding boxes corresponding to one or more objects within one or more digital images.

Abstract: Apparatuses, systems, and techniques to generate bounding box information. In at least one embodiment, for example, bounding box information is generated based, at least in part, on a plurality of candidate bounding box information.

Abstract: Apparatuses, systems, and techniques to train a machine-learned model. In at least one embodiment, a plurality of training clients each obtain an exclusive right to update a model in turn, and each client trains said model with training data not accessible to other training clients.

Abstract: A technique of implementing on-screen gestures associated with a software application comprises receiving a first control input that relates to a first scene associated with the software application, translating the first control input into a first set of instructions based on a first mapping, and providing the first set of instructions to an operating system that includes the first set of instructions in the software application, receiving a second control input that relates to a second scene associated with the software application, translating the second control input into a second set of instructions based on a second mapping, and providing the second set of instructions to the operating system, wherein the operating system is configured to include the second set of instructions in the software application.

Abstract: A method for producing a single-mode fiber for submarine cables including washing and flame polishing of a preform, fiber drawing, coating with a first coating layer and curing therein using an UV-curing device, coating with a second coating layer and curing therein using the UV-curing device, fiber selection with a 2% screening strain, and testing the properties of the fiber. The fiber has a high strength and long coiling length exceeding 100 km and the method is easy to practice with low production cost and parameters involved therein are highly controllable.

Abstract: A method for producing a single-mode fiber for submarine cables including washing and flame polishing of a preform, fiber drawing, coating with a first coating layer and curing therein using an UV-curing device, coating with a second coating layer and curing therein using the UV-curing device, fiber selection with a 2% screening strain, and testing the properties of the fiber. The fiber has a high strength and long coiling length exceeding 100 km and the method is easy to practice with low production cost and parameters involved therein are highly controllable.