Clustering, explainability, and automated decisions in computer-based reasoning systems

Patent metadata

Field	Value
Publication number	US-11823080-B2
Application number	US-202217962820-A
Country	US
Kind code	B2
Filing date	Oct 10, 2022
Priority date	Oct 30, 2018
Publication date	Nov 21, 2023
Grant date	Nov 21, 2023

Abstract

The techniques herein include using an input context to determine a suggested action and/or cluster. Explanations may also be determined and returned along with the suggested action. The explanations may include (i) one or more most similar cases to the suggested case (e.g., the case associated with the suggested action) and, optionally, a conviction score for each nearby cases; (ii) action probabilities, (iii) excluding cases and distances, (iv) archetype and/or counterfactual cases for the suggested action; (v) feature residuals; (vi) regional model complexity; (vii) fractional dimensionality; (viii) prediction conviction; (ix) feature prediction contribution; and/or other measures such as the ones discussed herein, including certainty. The explanation data may be used to determine whether to perform a suggested action.

First claim

What is claimed is:

1. A method comprising:
receiving a request for a suggested action based on an input context, in a computer-based reasoning system, wherein the computer-based reasoning system includes a computer-based reasoning model, wherein the input context and computer-based reasoning system relate to a controllable system, wherein the controllable system is one of a self-driving vehicle, an image labelling, a manufacturing control system, a smart voice control system, a federated device control system, an experimental control system, an energy transfer control system, a health care decision system, a health care fraud system, a financial decision system, and a financial fraud protection system, a cybersecurity control system;
determining one or more candidate cases based on the input context in the computer-based reasoning system, wherein the one or more candidate cases include respective one or more candidate actions;
determining the suggested action based on the respective one or more candidate actions;
determining a certainty score based on the suggested action, wherein the certainty score is determined based on a ratio between two different conviction scores each associated with the one or more candidate cases;
determining a regional computer-based reasoning model of two or more cases in the computer-based reasoning model near the one or more candidate cases, and wherein:
determining a first conviction score of the two different conviction scores each associated with the one or more candidate cases comprises performing a first conviction function associated with the regional computer-based reasoning model comprising:
removing a suggested case associated with the suggested action from the regional computer-based reasoning model;
adding the suggested case into the regional computer-based reasoning model,
wherein the first conviction function is a measure of how much information the suggested case distorts the regional computer-based reasoning model;

when the certainty score is not beyond a certain threshold:
determining one or more explanation factors for the suggested action determined based at least in part on the input context;
responding to the request for the suggested action with the suggested action and the one or more explanation factors,
wherein determining the suggested action comprises determining a cluster associated with each of the one or more candidate cases, determining a suggested cluster based on the cluster associated with each of the one or more candidate cases, and the suggested cluster is returned as the suggested action;
when the certainty score is beyond the certain threshold, using the input context and the suggested action for determining a contextually-determined action to be performed by the self-driving vehicle, a suggested label by the image labelling, a suggested manufacturing action to be performed by the manufacturing control system, a suggested smart voice action to be performed by the smart voice control system, a suggested federated device action to be performed by the federated device control system, a suggested experimental control to be performed by the experimental control system, a suggested energy transfer control action to be performed by the energy transfer control system, a suggested health care action to be performed by the health care decision system, a suggested health care fraud decision action to be performed by the health care fraud system, a suggested financial decision action to be performed by the financial decision system, and suggested financial fraud protection action to be performed by the financial fraud protection system, a suggested cybersecurity control action to be performed by the cybersecurity control system,
wherein the method is performed by one or more computing devices.

2. The method of claim 1 , wherein determining a second conviction score of the two different conviction scores each associated with the one or more candidate cases comprises performing a second conviction function associated with:
removing a suggested case associated with the suggested action from the computer-based reasoning model;
adding the suggested case into the computer-based reasoning model,
wherein the second conviction function is a measure of how much information the suggested case distorts the computer-based reasoning model.

3. The method of claim 1 , wherein determining a second conviction score of the two different conviction scores each associated with the one or more candidate cases comprises performing a second conviction function associated with:
removing a suggested case associated with the suggested action from the computer-based reasoning model; and
adding the suggested case into the computer-based reasoning model,
wherein the second conviction function is a measure of how much information is required to describe a position of the suggested case relative to existing cases in the computer-based reasoning model.

4. The method of claim 1 , wherein determining a second conviction score of the two different conviction scores each associated with the one or more candidate cases comprises performing a second conviction function associated with:
removing a suggested case associated with the suggested action from the computer-based reasoning model; and
adding the suggested case into the computer-based reasoning model,
wherein the conviction function is a measure of how much information is required to describe a position of the suggested case relative to existing cases in the computer-based reasoning model.

5. The method of claim 1 , comprising:
determining a regional computer-based reasoning model of two or more cases in the computer-based reasoning model near the one or more candidate cases, and wherein:
determining a second conviction score of the two different conviction scores each associated with the one or more candidate cases comprises performing a second conviction function associated with the regional computer-based reasoning model comprising:
removing a suggested case associated with the suggested action from the regional computer-based reasoning model; and
adding the suggested case into the regional computer-based reasoning model,
wherein the second conviction function is a measure of how much information is required to describe a position of the suggested case relative to existing cases in the regional computer-based reasoning model.

6. A system for performing a machine-executed operation involving instructions, wherein said instructions are instructions which, when executed by one or more computing devices, cause performance of a process comprising:
receiving a request for a suggested action based on an input context, in a computer-based reasoning system, wherein the computer-based reasoning system includes a computer-based reasoning model, wherein the input context and computer-based reasoning system relate to a controllable system, wherein the controllable system is one of a self-driving vehicle, an image labelling, a manufacturing control system, a smart voice control system, a federated device control system, an experimental control system, an energy transfer control system, a health care decision system, a health care fraud system, a financial decision system, and a financial fraud protection system, a cybersecurity control system;
determining one or more candidate cases based on the input context in the computer-based reasoning system, wherein the one or more candidate cases include respective one or more candidate actions;
determining the suggested action based on the respective one or more candidate actions;
determining a certainty score based on the suggested action, wherein the certainty score is determined based on a ratio between two different conviction scores each associated with

Assignees

• Diveplane Corp

Inventors

• Hazard Christopher James
• Resnick Michael
• Fusting Christopher

Classifications

• G06N5/04 Primary

  Inference or reasoning models · CPC title

• G06N5/045 Primary

  Explanation of inference; Explainable artificial intelligence [XAI]; Interpretable artificial intelligence · CPC title

• G06N3/02

  Neural networks · CPC title

• G06N5/025

  Extracting rules from data · CPC title

• G06N20/00

  Machine learning · CPC title