The present disclosure relates generally to techniques for enhancing recurrent neural networks (RNNs) implemented on an integrated circuit. In particular, approximations of activation functions used in an RNN, such as sigmoid and hyperbolic tangent, may be implemented in an integrated circuit, which may result in increased efficiencies, reduced latency, increased accuracy, and reduced resource consumption involved with implementing machine learning.

The present disclosure relates generally to techniques for enhancing recurrent neural networks (RNNs) implemented on an integrated circuit. In particular, approximations of activation functions used in an RNN, such as sigmoid and hyperbolic tangent, may be implemented in an integrated circuit, which may result in increased efficiencies, reduced latency, increased accuracy, and reduced resource consumption involved with implementing machine learning.

Systems methods and media for preference clustering are provided. In one example, a clustering system for analyzing a cluster comprises processors and a memory storing instructions that cause the system to calculate a Distance Angular Measure (DAM) for the cluster, the (DAM) comprising a distance component and an angular component of the cluster. In one example, the distance component of the (DAM) includes one of a cluster variation and a cluster radius.

Systems methods and media for preference clustering are provided. In one example, a clustering system for analyzing a cluster comprises processors and a memory storing instructions that cause the system to calculate a Distance Angular Measure (DAM) for the cluster, the (DAM) comprising a distance component and an angular component of the cluster. In one example, the distance component of the (DAM) includes one of a cluster variation and a cluster radius.

The embodiments disclosed herein include improved toric lenses and other ophthalmic apparatuses (including, for example, contact lens, intraocular lenses (IOLs), and the like) and associated method for their design and use. In an embodiment, an ophthalmic apparatus (e.g., a toric lens) includes one or more angularly-varying phase members comprising a diffractive or refractive structure, each varying the depths of focus of the apparatus so as to provide an extended tolerance to misalignment of the apparatus when implanted in an eye. That is, the ophthalmic apparatus establishes an extended band of operational meridian over the intended correction meridian.

Embodiments of systems, apparatuses, and methods for performing vector-packed controllable sine and/or cosine operations in a processor are described. For example, execution circuitry executes a decoded instruction to compute at least a real output value and an imaginary output value based on at least a cosine calculation and a sine calculation, the cosine and sine calculations each based on an index value from a packed data source operand, add the index value with an index increment value from the packed data source operand to create an updated index value, and store the real output value, the imaginary output value, and the updated index value to a packed data destination operand.

Embodiments of systems, apparatuses, and methods for performing vector-packed controllable sine and/or cosine operations in a processor are described. For example, execution circuitry executes a decoded instruction to compute at least a real output value and an imaginary output value based on at least a cosine calculation and a sine calculation, the cosine and sine calculations each based on an index value from a packed data source operand, add the index value with an index increment value from the packed data source operand to create an updated index value, and store the real output value, the imaginary output value, and the updated index value to a packed data destination operand.

According to one embodiment, provided is a chip recognition system that recognizes a chip on a gaming table in an amusement place having the gaming table, the chip recognition system including: a game recording apparatus that records, as an image, a state of chips stacked on the gaming table, using a camera; an image analysis apparatus that performs an image analysis on the recorded image of the state of chips; a plurality of chip determination apparatuses including at least a first artificial intelligence apparatus that determines a number of the chips stacked, using an image analysis result obtained by the image analysis apparatus; and a second artificial intelligence apparatus that decides a correct number of the chips stacked, when the plurality of chip determination apparatuses obtain different determination results for the number of the chips stacked.

According to one embodiment, provided is a chip recognition system that recognizes a chip on a gaming table in an amusement place having the gaming table, the chip recognition system including: a game recording apparatus that records, as an image, a state of chips stacked on the gaming table, using a camera; an image analysis apparatus that performs an image analysis on the recorded image of the state of chips; a plurality of chip determination apparatuses including at least a first artificial intelligence apparatus that determines a number of the chips stacked, using an image analysis result obtained by the image analysis apparatus; and a second artificial intelligence apparatus that decides a correct number of the chips stacked, when the plurality of chip determination apparatuses obtain different determination results for the number of the chips stacked.

The present disclosure relates generally to techniques for enhancing recurrent neural networks (RNNs) implemented on an integrated circuit. In particular, approximations of activation functions used in an RNN, such as sigmoid and hyperbolic tangent, may be implemented in an integrated circuit, which may result in increased efficiencies, reduced latency, increased accuracy, and reduced resource consumption involved with implementing machine learning.