Abstract: Embodiments are generally directed to techniques for accelerating neural networks. Many embodiments include a hardware accelerator for a bi-directional multi-layered GRU and LC neural network. Some embodiments are particularly directed to a hardware accelerator that enables offloading of the entire LC+GRU network to the hardware accelerator. Various embodiments include a hardware accelerator with a plurality of matrix vector units to perform GRU steps in parallel with LC steps. For example, at least a portion of computation by a first matrix vector unit of a GRU step in a neural network may overlap at least a portion of computation by a second matrix vector unit of an output feature vector for the neural network. Several embodiments include overlapping computation associated with a layer of a neural network with data transfer associated with another of the neural network.

Abstract: To view media, a user may select a media stream by operating a portable device that controls a media presentation device. The portable device may be configured to multicast this stream selection to both the media presentation device and a selection analysis machine. The remote control may have or include both an infrared emitter and a cellular telephone, and the stream selection may be sent both to the media presentation device and to the selection analysis machine. The selection analysis machine may receive and store stream selections over a period of time, and these aggregated stream selections may form all or part of a profile of a user or a group of users who use the media presentation device. This profile may indicate viewing habits and choices of one or more users of the media presentation device, and the selection analysis machine may analyze this profile.