US 12,430,072 B2
Storage controller having data augmentation components for use with non-volatile memory die
Alexander Bazarsky, Holon (IL); and Ariel Navon, Revava (IL)
Assigned to Sandisk Technologies, Inc., Milpitas, CA (US)
Filed by Western Digital Technologies, Inc., San Jose, CA (US)
Filed on Aug. 26, 2022, as Appl. No. 17/897,040.
Application 17/897,040 is a continuation of application No. 16/718,148, filed on Dec. 17, 2019, granted, now 11,520,521.
Application 16/718,148 is a continuation in part of application No. 16/447,619, filed on Jun. 20, 2019, granted, now 11,501,109.
Prior Publication US 2022/0413769 A1, Dec. 29, 2022
Int. Cl. G06F 3/06 (2006.01); G06F 11/07 (2006.01); G06F 11/10 (2006.01); G06K 9/00 (2022.01); G06N 20/00 (2019.01); G06T 3/20 (2006.01); G06T 3/608 (2024.01); G06T 5/70 (2024.01); G11B 27/36 (2006.01)
CPC G06F 3/0659 (2013.01) [G06F 3/0604 (2013.01); G06F 3/0688 (2013.01); G06F 11/076 (2013.01); G06F 11/1044 (2013.01); G06N 20/00 (2019.01); G06T 3/20 (2013.01); G06T 3/608 (2013.01); G06T 5/70 (2024.01); G06T 2207/10004 (2013.01); G06T 2207/20172 (2013.01)] 24 Claims
OG exemplary drawing
 
1. A data storage device, comprising:
a memory die with non-volatile memory (NVM) elements; and
a non-transitory storage controller that is separate from the memory die, the storage controller configured to:
obtain machine learning training images from the NVM elements of the memory die, and
augment the machine learning training images by being further configured to: (a) generate altered machine learning training images using an image augmentation component of the storage controller and (b) control the memory die to generate other altered machine learning training images for retrieval by the storage controller;
wherein the storage controller comprises a machine learning controller configured to train a deep learning accelerator (DLA) using the altered machine learning training images.