Abstract: Methods and apparatus are disclosed herein for computation and compression efficiency in distributed video analytics. Example apparatus disclosed herein are to identify a key frame and a non-key frame in a video frame sequence input to a neural network at a client server, determine motion information between the key frame and the non-key frame based on optical flow, and determine a frame feature representation based on the motion information reconstructed at an edge server, the motion information including feature warping residual errors.

Abstract: One embodiment provides a graphics processor comprising a block of execution resources, a cache memory, a cache memory prefetcher, and circuitry including a programmable neural network unit, the programmable neural network unit comprising a network hardware block including circuitry to perform neural network operations and activation operations for a layer of a neural network, the programmable neural network unit addressable by cores within the block of graphics cores and the neural network hardware block configured to perform operations associated with a neural network configured to determine a prefetch pattern for the cache memory prefetcher.

Abstract: One embodiment provides for a graphics processor comprising a block of graphics compute units, a graphics processor pipeline coupled to the block of graphics compute units, and a programmable neural network unit including one or more neural network hardware blocks. The programmable neural network unit is coupled with the block of graphics compute units and the graphics processor pipeline. The one or more neural network hardware blocks include hardware to perform neural network operations and activation operations for a layer of a neural network. The programmable neural network unit can configure settings of one or more hardware blocks within the graphics processor pipeline based on a machine learning model trained to optimize performance of a set of workloads.

Abstract: Technologies for dynamic wireless noise mitigation include a computing device having a wireless modem and one or more antennas. The computing device activates one or more components of the computing device, monitors platform activity, and measures wireless noise received by the antennas. The computing device trains a noise prediction model based on the platform activity and the measured noise. The computing device may monitor platform activity and predict a noise prediction with the noise prediction model based on the monitored activity. The computing device may mitigate wireless noise received by the wireless antennas based on the noise prediction. The computing device may provide the noise prediction to the wireless modem. Other embodiments are described and claimed.

Abstract: One embodiment provides for a graphics processor comprising a block of graphics compute units, a graphics processor pipeline coupled to the block of graphics compute units, and a programmable neural network unit including one or more neural network hardware blocks. The programmable neural network unit is coupled with the block of graphics compute units and the graphics processor pipeline. The one or more neural network hardware blocks include hardware to perform neural network operations and activation operations for a layer of a neural network. The programmable neural network unit can configure settings of one or more hardware blocks within the graphics processor pipeline based on a machine learning model trained to optimize performance of a set of workloads.

Abstract: One embodiment provides for a graphics processor comprising a block of graphics compute units, a graphics processor pipeline coupled to the block of graphics compute units, and a programmable neural network unit including one or more neural network hardware blocks. The programmable neural network unit is coupled with the block of graphics compute units and the graphics processor pipeline. The one or more neural network hardware blocks include hardware to perform neural network operations and activation operations for a layer of a neural network. The programmable neural network unit can configure settings of one or more hardware blocks within the graphics processor pipeline based on a machine learning model trained to optimize performance of a set of workloads.

Abstract: Technologies for dynamic wireless noise mitigation include a computing device having a wireless modem and one or more antennas. The computing device activates one or more components of the computing device, monitors platform activity, and measures wireless noise received by the antennas. The computing device trains a noise prediction model based on the platform activity and the measured noise. The computing device may monitor platform activity and predict a noise prediction with the noise prediction model based on the monitored activity. The computing device may mitigate wireless noise received by the wireless antennas based on the noise prediction. The computing device may provide the noise prediction to the wireless modem. Other embodiments are described and claimed.

Abstract: One embodiment provides for a graphics processor comprising a block of graphics compute units, a graphics processor pipeline coupled to the block of graphics compute units, and a programmable neural network unit including one or more neural network hardware blocks. The programmable neural network unit is coupled with the block of graphics compute units and the graphics processor pipeline. The one or more neural network hardware blocks include hardware to perform neural network operations and activation operations for a layer of a neural network. The programmable neural network unit can configure settings of one or more hardware blocks within the graphics processor pipeline based on a machine learning model trained to optimize performance of a set of workloads.

Abstract: Some demonstrative embodiments include devices, systems of selectively providing Internet Protocol (IP) session continuity. In one example, a mobile device may include a radio to communicate with a wireless network, the radio to transmit a session setup request to setup a communication session, and to receive a session setup response in response to the session setup request, the session setup response including a first Internet Protocol (IP) address and a second IP address assigned to the communication session, and an indication that the first IP address is configured to maintain IP session continuity; and a controller to select to use the first IP address for the communication session, if IP session continuity is to be maintained for the communication session, and to select to use the second IP address for the communication session, if IP session continuity is not to be maintained for the communication session.

Abstract: Some demonstrative embodiments include devices, systems of selectively providing Internet Protocol (IP) session continuity. In one example, a mobile device may include a radio to communicate with a wireless network, the radio to transmit a session setup request to setup a communication session, and to receive a session setup response in response to the session setup request, the session setup response including a first Internet Protocol (IP) address and a second IP address assigned to the communication session, and an indication that the first IP address is configured to maintain IP session continuity; and a controller to select to use the first IP address for the communication session, if IP session continuity is to be maintained for the communication session, and to select to use the second IP address for the communication session, if IP session continuity is not to be maintained for the communication session.