Ethical confidence fabrics: measuring ethical algorithm development

Patent metadata

Field	Value
Publication number	US-12591827-B2
Application number	US-202217648629-A
Country	US
Kind code	B2
Filing date	Jan 21, 2022
Priority date	Jan 21, 2022
Publication date	Mar 31, 2026
Grant date	Mar 31, 2026

Abstract

One example method includes formulating a hypothesis for development of computing model, annotating the hypothesis with ethics metadata, storing the hypothesis and the ethics metadata, in association with each other, in a ledger, performing ‘n’ phases of a development lifecycle for the computing model, annotating each of the ‘n’ phases with ethics metadata specific to the phase, updating the ledger to include the ‘n’ phases and the ethics metadata respectively associated with each of the ‘n’ phases, and calculating an ethics confidence score for the computing model.

First claim

What is claimed is:

1 . A method, comprising:
formulating a hypothesis for development of a computing model; annotating the hypothesis with first ethics metadata; storing the hypothesis and annotations regarding the first ethics metadata, in association with each other, in a ledger; performing ‘n’ phases of a development lifecycle for the computing model, wherein ‘n’ is greater than or equal to four; annotating each of the ‘n’ phases with second ethics metadata specific to the phase; updating the ledger to include the ‘n’ phases and annotations regarding the second ethics metadata respectively associated with each of the ‘n’ phases; and calculating an ethics confidence score for the computing model and/or for the development lifecycle based on the annotations of the hypothesis regarding the first ethics metadata and the annotations of each of the ‘n’ phases regarding the second ethic metadata, wherein the ‘n’ phases include a data discovery phase, in which data that helps prove the hypothesis is discovered, and wherein common ethics metadata is included in the first ethics metadata and the second ethics metadata.

2 . The method as recited in claim 1 , wherein the ‘n’ phases further include any one or more of: a data preparation phase; a model planning phase; a model building phase; a results communication phase; and, a model deployment phase.

3 . The method as recited in claim 1 , wherein the second ethics metadata for one or more of the ‘n’ phases comprises ethics metadata supplied by an enterprise that is developing the computing model and/or ethics metadata supplied by a trusted third party.

4 . The method as recited in claim 1 , wherein the computing model comprises a data analysis model.

5 . The method as recited in claim 1 , further comprising storing the ethics confidence score in the ledger.

6 . The method as recited in claim 1 , wherein the ethics confidence score is calculated based on the respective annotations of one or more of the ‘n’ phases.

7 . The method as recited in claim 1 , further comprising storing the model, the annotations regarding the first ethics metadata, the annotations regarding the second ethics metadata, the ethics confidence score, and an equation used to calculate the ethics confidence score, in the ledger.

8 . The method as recited in claim 1 , further comprising modifying, based on the ethics confidence score, one of the ‘n’ phases of the development lifecycle for the computing model.

9 . The method as recited in claim 1 , further comprising:
deploying the computing model in a production environment when the ethics confidence score meets or exceeds a threshold value, wherein the computing model is not deployed in the production environment when the ethics confidence score falls below the threshold value.

10 . The method as recited in claim 1 , wherein the annotating of the hypothesis and the annotating of the ‘n’ phases are performed using an ethics confidence fabric.

11 . A non-transitory storage medium having stored therein instructions that are executable by one or more hardware processors to perform operations comprising:
formulating a hypothesis for development of a computing model; annotating the hypothesis with first ethics metadata; storing the hypothesis and annotations regarding the first ethics metadata, in association with each other, in a ledger; performing ‘n’ phases of a development lifecycle for the computing model, wherein ‘n’ is greater than or equal to four; annotating each of the ‘n’ phases with second ethics metadata specific to the phase; updating the ledger to include the ‘n’ phases and annotations regarding the second ethics metadata respectively associated with each of the ‘n’ phases; and calculating an ethics confidence score for the computing model and/or for the development lifecycle based on the annotations of the hypothesis regarding the first ethics metadata and the annotations of each of the ‘n’ phases regarding the second ethic metadata, wherein the ‘n’ phases include a data discovery phase, in which data that helps prove the hypothesis is discovered, and wherein common ethics metadata is included in the first ethics metadata and the second ethics metadata.

12 . The non-transitory storage medium as recited in claim 11 ,
wherein the ‘n’ phases further include any one or more of: a data preparation phase; a model planning phase; a model building phase; a results communication phase; and, a model deployment phase.

13 . The non-transitory storage medium as recited in claim 11 , wherein the second ethics metadata for one or more of the ‘n’ phases comprises ethics metadata supplied by an enterprise that is developing the computing model and/or ethics metadata supplied by a trusted third party.

14 . The non-transitory storage medium as recited in claim 11 , wherein the computing model comprises a data analysis model.

15 . The non-transitory storage medium as recited in claim 11 , wherein the operations further comprise storing the ethics confidence score in the ledger.

16 . The non-transitory storage medium as recited in claim 11 , wherein the ethics confidence score is calculated based on the respective annotations of one or more of the ‘n’ phases.

17 . The non-transitory storage medium as recited in claim 11 , wherein the operations further comprise storing the model, the annotations regarding the first ethics metadata, the annotations regarding the second ethics metadata, the ethics confidence score, and an equation used to calculate the ethics confidence score, in the ledger.

18 . The non-transitory storage medium as recited in claim 11 , wherein the operations further comprise modifying, based on the ethics confidence score, one of the ‘n’ phases of the development lifecycle for the computing model.

19 . The non-transitory storage medium as recited in claim 11 , wherein the operations further comprise deploying the computing model in a production environment when the ethics confidence score meets or exceeds a threshold value,
wherein the computing model is not deployed in the production environment when the ethics confidence score falls below the threshold value.

20 . The non-transitory storage medium as recited in claim 11 , wherein the annotating of the hypothesis and the annotating of the ‘n’ phases are performed using an ethics confidence fabric.

Assignees

• Dell Products Lp

Inventors

• Reineke Nicole
• Todd Stephen J

Classifications

• G06N5/04

  Inference or reasoning models · CPC title

• G06N5/02

  Knowledge representation; Symbolic representation · CPC title

• G06Q10/06393

  Score-carding, benchmarking or key performance indicator [KPI] analysis · CPC title

• G06Q10/067

  Enterprise or organisation modelling · CPC title

• G06N5/022

  Knowledge engineering; Knowledge acquisition · CPC title