Abstract: A three dimensional neural network accelerator that includes a first neural network accelerator tile that includes a first transmission coil, and a second neural network accelerator tile that includes a second transmission coil, wherein the first neural network accelerator tile is adjacent to and aligned vertically with the second neural network accelerator tile, and wherein the first transmission coil is configured to wirelessly communicate with the second transmission coil via inductive coupling.

Abstract: In one set of embodiments, a hardware module of a computer system can receive a stream of addresses corresponding to memory units being accessed by a central processing unit (CPU) of the computer system. The hardware module can further generate a frequency estimate for each address in the stream of addresses, the frequency estimate being indicative of a number of times a memory unit identified by the address has been accessed by the CPU, and can determine, based on the generated frequency estimates, a set of n most frequently accessed memory units.

Abstract: In one set of embodiments, a hardware module of a computer system can receive a stream of addresses corresponding to memory units being accessed by a central processing unit (CPU) of the computer system. The hardware module can further generate a frequency estimate for each address in the stream of addresses, the frequency estimate being indicative of a number of times a memory unit identified by the address has been accessed by the CPU, and can determine, based on the generated frequency estimates, a set of n most frequently accessed memory units.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for three-dimensionally stacked neural network accelerators. In one aspect, a method includes obtaining data specifying that a tile from a plurality of tiles in a three-dimensionally stacked neural network accelerator is a faulty tile. The three-dimensionally stacked neural network accelerator includes a plurality of neural network dies, each neural network die including a respective plurality of tiles, each tile has input and output connections. The three-dimensionally stacked neural network accelerator is configured to process inputs by routing the input through each of the plurality of tiles according to a dataflow configuration and modifying the dataflow configuration to route an output of a tile before the faulty tile in the dataflow configuration to an input connection of a tile that is positioned above or below the faulty tile on a different neural network die than the faulty tile.

Abstract: Described herein are systems, methods, and software to manage time calibration associated with an oscillator of a computing system. In one example, a computing system monitors clock cycles for an oscillator on the computing system, receives timing messages from a server, and calculates the frequency of the oscillator at intervals based on the monitored clock cycles and timing messages. The computing system further identifies a temperature from a temperature sensor at each of the intervals and generates a function to demonstrate frequency of the oscillator versus temperatures from the temperature sensor based on the identified temperatures and frequencies at the intervals.

Abstract: A three dimensional neural network accelerator that includes a first neural network accelerator tile that includes a first transmission coil, and a second neural network accelerator tile that includes a second transmission coil, wherein the first neural network accelerator tile is adjacent to and aligned vertically with the second neural network accelerator tile, and wherein the first transmission coil is configured to wirelessly communicate with the second transmission coil via inductive coupling.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for three-dimensionally stacked neural network accelerators. In one aspect, a method includes obtaining data specifying that a tile from a plurality of tiles in a three-dimensionally stacked neural network accelerator is a faulty tile. The three-dimensionally stacked neural network accelerator includes a plurality of neural network dies, each neural network die including a respective plurality of tiles, each tile has input and output connections. The three-dimensionally stacked neural network accelerator is configured to process inputs by routing the input through each of the plurality of tiles according to a dataflow configuration and modifying the dataflow configuration to route an output of a tile before the faulty tile in the dataflow configuration to an input connection of a tile that is positioned above or below the faulty tile on a different neural network die than the faulty tile.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for three-dimensionally stacked neural network accelerators. In one aspect, a method includes obtaining data specifying that a tile from a plurality of tiles in a three-dimensionally stacked neural network accelerator is a faulty tile. The three-dimensionally stacked neural network accelerator includes a plurality of neural network dies, each neural network die including a respective plurality of tiles, each tile has input and output connections. The three-dimensionally stacked neural network accelerator is configured to process inputs by routing the input through each of the plurality of tiles according to a dataflow configuration and modifying the dataflow configuration to route an output of a tile before the faulty tile in the dataflow configuration to an input connection of a tile that is positioned above or below the faulty tile on a different neural network die than the faulty tile.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for three-dimensionally stacked neural network accelerators. In one aspect, a method includes obtaining data specifying that a tile from a plurality of tiles in a three-dimensionally stacked neural network accelerator is a faulty tile. The three-dimensionally stacked neural network accelerator includes a plurality of neural network dies, each neural network die including a respective plurality of tiles, each tile has input and output connections. The three-dimensionally stacked neural network accelerator is configured to process inputs by routing the input through each of the plurality of tiles according to a dataflow configuration and modifying the dataflow configuration to route an output of a tile before the faulty tile in the dataflow configuration to an input connection of a tile that is positioned above or below the faulty tile on a different neural network die than the faulty tile.

Abstract: A three dimensional neural network accelerator that includes a first neural network accelerator tile that includes a first transmission coil, and a second neural network accelerator tile that includes a second transmission coil, wherein the first neural network accelerator tile is adjacent to and aligned vertically with the second neural network accelerator tile, and wherein the first transmission coil is configured to wirelessly communicate with the second transmission coil via inductive coupling.

Abstract: A three dimensional neural network accelerator that includes a first neural network accelerator tile that includes a first transmission coil, and a second neural network accelerator tile that includes a second transmission coil, wherein the first neural network accelerator tile is adjacent to and aligned vertically with the second neural network accelerator tile, and wherein the first transmission coil is configured to wirelessly communicate with the second transmission coil via inductive coupling.

Abstract: An optical security device can be used to view sensitive information provided in an obscured format via a potentially untrusted and/or compromised computer. The techniques described herein enable use of untrusted computers for access to sensitive information. The optical security device employs one or more forms of visual cryptography such as spatial cryptography and/or temporal cryptography in some instances via a programmable mask and/or a programmable color filter to reveal sensitive information that is provided in an obscured form by a potentially untrusted computer.

Abstract: An optical security device can be used to view sensitive information provided in an obscured format via a potentially untrusted and/or compromised computer. The techniques described herein enable use of untrusted computers for access to sensitive information. The optical security device employs one or more forms of visual cryptography such as spatial cryptography and/or temporal cryptography in some instances via a programmable mask and/or a programmable color filter to reveal sensitive information that is provided in an obscured form by a potentially untrusted computer.

Abstract: A reduced-latency ink rendering system and method that reduces latency in rendering ink on a display by bypassing at least some layers of the operating system. “Ink” is any input from a user through a touchscreen device using the user's finger or a pen. Moreover, some embodiments of the system and method avoid the operating system and each central-processing unit (CPU) on a computing device when initially rendering the ink by going directly from the digitizer to the display controller. Any correction or additional processing of the rendered ink is performed after the initial rendering of the ink. Embodiments of the system and method address ink-rendering latency in software embodiments, which include techniques to bypass the typical rendering pipeline and quickly render ink on the display, and hardware embodiments, which use hardware and techniques that locally change display pixels. These embodiments can be mixed and matched in any manner.

Abstract: An optical security device can be used to view sensitive information provided in an obscured format via a potentially untrusted and/or compromised computer. The techniques described herein enable use of untrusted computers for access to sensitive information. The optical security device employs one or more forms of visual cryptography such as spatial cryptography and/or temporal cryptography in some instances via a programmable mask and/or a programmable color filter to reveal sensitive information that is provided in an obscured form by a potentially untrusted computer.

Abstract: An optical security device can be used to view sensitive information provided in an obscured format via a potentially untrusted and/or compromised computer. The techniques described herein enable use of untrusted computers for access to sensitive information. The optical security device employs one or more forms of visual cryptography such as spatial cryptography and/or temporal cryptography in some instances via a programmable mask and/or a programmable color filter to reveal sensitive information that is provided in an obscured form by a potentially untrusted computer.

Abstract: A reduced-latency ink rendering system and method that reduces latency in rendering ink on a display by bypassing at least some layers of the operating system. “Ink” is any input from a user through a touchscreen device using the user's finger or a pen. Moreover, some embodiments of the system and method avoid the operating system and each central-processing unit (CPU) on a computing device when initially rendering the ink by going directly from the digitizer to the display controller. Any correction or additional processing of the rendered ink is performed after the initial rendering of the ink. Embodiments of the system and method address ink-rendering latency in software embodiments, which include techniques to bypass the typical rendering pipeline and quickly render ink on the display, and hardware embodiments, which use hardware and techniques that locally change display pixels. These embodiments can be mixed and matched in any manner.

Abstract: A multi-purpose stylus and method for communicating and interacting with a computing device both through physical contact and wirelessly. Embodiments of the stylus and method facilitate the use of the stylus as both a physical input instrument (by contacting a surface of the computing device to input data) and as a remote wireless instrument using a variety of auxiliary devices. Many types and combinations of auxiliary devices may be incorporated into embodiments of the stylus and method. These include one or more microphones and speakers, a laser pointer, a camera, a color sensor for obtaining color coordinates of an object, and an accelerometer to identify and interpret user gestures. Moreover, identification and authentication of a user may be achieved by including a fingerprint sensor and an identification device having a unique identifier. A transceiver is used to wireless communicate and control remote devices and the computing device.

Abstract: In a system for communicating optically encoded data among a network of nodes, the nodes are connected to each other by optical transmit and receive fibers. Each node includes a laser, a transmitter, and a receiver connected to the transmitters of the other nodes by a fiber stretcher. The rate at which the laser generates light pulses is controlled by a frequency control signal supplied to an oscillator connected to the laser. The delay from the transmitters to the receivers is controlled by a phase control signal supplied to the fiber stretcher connected to the receive fiber. The transmitter connected to the transmit fiber selective delays the pulses into a plurality of time multiplexed channels. The receiver connected to the fiber stretcher selectively detects the pulses in the time multiplexed channels while all of the pulses of all of the nodes are globally synchronized in all of the time multiplexed channels in frequency and phase using the frequency and phase control signals.