公开(公告)号：US11610105B2

公开(公告)日：2023-03-21

申请号：US16386849

申请日：2019-04-17

Applicant: Hewlett Packard Enterprise Development LP

Inventor： Suhas Kumar ,  Thomas Van Vaerenbergh ,  John Paul Strachan

IPC: G06N3/06 ,  G06N3/04 ,  G06N3/063 ,  G11C13/00 ,  H03K3/0233

Abstract: Systems and methods are provided for implementing a hardware accelerator. The hardware accelerator emulates a neural network, and includes a memristor crossbar array, and a non-linear filter. The memristor crossbar array can be programmed to calculate node values of the neural network. The nodes values can be calculated in accordance with rules to reduce an energy function associated with the neural network. The non-linear filter is coupled to the memristor crossbar array and programmed to harness noise signals that may be present in analog circuitry of the hardware accelerator. The noise signals can be harnessed such that the energy function associated with the neural network converges towards a global minimum and modifies the calculated node values. In some embodiments, the non-liner filter is implemented as a Schmidt trigger comparator.

公开(公告)号：US20200334523A1

公开(公告)日：2020-10-22

申请号：US16386849

申请日：2019-04-17

Applicant: Hewlett Packard Enterprise Development LP

Inventor： Suhas Kumar ,  Thomas Van Vaerenbergh ,  John Paul Strachan

IPC: G06N3/063 ,  H03K3/0233 ,  G11C13/00 ,  G06N3/04

Abstract: Systems and methods are provided for implementing a hardware accelerator. The hardware accelerator emulates a neural network, and includes a memristor crossbar array, and a non-linear filter. The memristor crossbar array can be programmed to calculate node values of the neural network. The nodes values can be calculated in accordance with rules to reduce an energy function associated with the neural network. The non-linear filter is coupled to the memristor crossbar array and programmed to harness noise signals that may be present in analog circuitry of the hardware accelerator. The noise signals can be harnessed such that the energy function associated with the neural network converges towards a global minimum and modifies the calculated node values. In some embodiments, the non-liner filter is implemented as a Schmidt trigger comparator.