| CPC G06F 13/4027 (2013.01) [G06F 7/50 (2013.01); G06F 13/1668 (2013.01); G06F 17/16 (2013.01); G06N 20/00 (2019.01); G11C 11/22 (2013.01); H01L 25/0657 (2013.01); G06N 3/063 (2013.01)] | 18 Claims |

|
1. A method comprising:
loading a first input and a second input from a first die to a second die, wherein the first die is on a substrate, wherein the first die is between the substrate and the second die, wherein the first die is a memory die, wherein the second die is a compute die, wherein the memory die comprises non-volatile memory, wherein the second die is between a heat sink and the first die; wherein the substrate is at a reference level of an x-y plane, wherein the first die is positioned above the reference level along a positive z-axis at a first z-plane, wherein the second die is positioned above the first z-plane along the positive z-axis at a second z-plane, and wherein the second z-plane is higher than the first z-plane along an x-axis and relative to the reference level;
loading a first weight and a second weight from the first die to the second die;
performing a first multiplication of the first input and the first weight;
performing a second multiplication of the second input and the second weight; and
generating an output according to outputs from the first multiplication and the second multiplication, wherein the second die is over the first die along the positive z-axis.
|