Title:
HIERARCHICAL METHODS AND SYSTEMS FOR STORING DATA
Document Type and Number:
WIPO Patent Application WO/2021/223098
Kind Code:
A1
Abstract:
Disclosed are systems and methods that determine whether instances of data (e.g., forward activations, backward derivatives of activations) that are used to train deep neural networks are to be stored on-chip or off-chip. The disclosed systems and methods are also used to prune the data (discard or delete selected instances of data). A system includes a hierarchical arrangement of on-chip and off-chip memories, and also includes a hierarchical arrangement of data selector devices that are used to decide whether to discard data and where in the system the data is to be discarded.
Inventors:
QIN MINGHAI (US)
LIU CHUNSHENG (US)
XIAO ZHIBIN (US)
GUAN TIANCHAN (CN)
GAO YUAN (CN)
LIU CHUNSHENG (US)
XIAO ZHIBIN (US)
GUAN TIANCHAN (CN)
GAO YUAN (CN)
Application Number:
PCT/CN2020/088761
Publication Date:
November 11, 2021
Filing Date:
May 06, 2020
Export Citation:
Assignee:
ALIBABA GROUP HOLDING LTD (CN)
International Classes:
G06F13/16; G11C14/00
Foreign References:
| US20200034318A1 | 2020-01-30 | |||
| CN110837481A | 2020-02-25 | |||
| CN107844321A | 2018-03-27 | |||
| CN101995547A | 2011-03-30 | |||
| CN111078592A | 2020-04-28 | |||
| US20200110990A1 | 2020-04-09 | |||
| US20130191587A1 | 2013-07-25 | |||
| US20200133874A1 | 2020-04-30 |
Attorney, Agent or Firm:
BEIJING TSINGYUANHUI INTELLECTUAL PROPERTY LAW FIRM (CN)
Download PDF: