US 11,810,008 B2
Data blaming
William Redington Hewlett, II, Mountain View, CA (US); Seokkyung Chung, Sunnyvale, CA (US); and Lin Xu, Saratoga, CA (US)
Assigned to Palo Alto Networks, Inc., Santa Clara, CA (US)
Filed by Palo Alto Networks, Inc., Santa Clara, CA (US)
Filed on Aug. 6, 2022, as Appl. No. 17/882,555.
Application 17/882,555 is a continuation of application No. 16/357,064, filed on Mar. 18, 2019, granted, now 11,455,551.
Application 16/357,064 is a continuation of application No. 14/810,335, filed on Jul. 27, 2015, granted, now 10,296,836, issued on May 21, 2019.
Claims priority of provisional application 62/141,202, filed on Mar. 31, 2015.
Prior Publication US 2022/0374724 A1, Nov. 24, 2022
This patent is subject to a terminal disclaimer.
Int. Cl. G06N 5/04 (2023.01); G06N 20/00 (2019.01); G06N 20/20 (2019.01); G06F 18/2431 (2023.01); G06N 20/10 (2019.01); G06N 5/01 (2023.01); G06N 5/025 (2023.01)
CPC G06N 5/04 (2013.01) [G06F 18/2431 (2023.01); G06N 20/00 (2019.01); G06N 20/20 (2019.01); G06N 5/01 (2023.01); G06N 5/025 (2013.01); G06N 20/10 (2019.01)] 11 Claims
OG exemplary drawing
 
1. A system, comprising:
a processor configured to:
receive a copy of a model comprising a plurality of trees and receive a copy of training set data comprising a plurality of training set examples;
for each tree included in the plurality of trees, use the training set data to determine which of the training set examples included in the plurality of training set examples are classified as a given leaf, and
generate a blame forest at least in part by mapping each training set item to the respective leaves at which it arrives; and
a memory coupled to the processor and configured to provide the processor with instructions.