IBM Patent US12586683B2 for Selective Label Decision-Making
Summary
The USPTO has granted IBM patent US12586683B2 for a computer-implemented method of decision-making using selective labels in health informatics. The patent describes a system for rendering decisions based on conditional success probabilities and confidence values.
What changed
The United States Patent and Trademark Office (USPTO) has granted International Business Machines Corporation (IBM) patent US12586683B2, titled "Decision-making under selective labels." This patent, filed on July 20, 2021, and granted on March 24, 2026, covers a computer-implemented method for decision-making using selective labels, particularly relevant to health informatics. The method involves receiving conditional success probability values and confidence values associated with a feature of an entity, selecting a parameter that balances short-term learning with long-term utility, and rendering a decision based on a machine learning policy.
This patent grant is primarily of informational value, as it represents intellectual property rather than a regulatory mandate. However, it highlights advancements in AI and machine learning within the healthcare sector. Compliance officers in health informatics or technology development should be aware of such patented technologies, as they may influence future product development, competitive landscapes, and potential licensing considerations. No immediate compliance actions are required, but awareness of IBM's patented methods in selective label decision-making is recommended for strategic planning.
Source document (simplified)
Decision-making under selective labels
Grant US12586683B2 Kind: B2 Mar 24, 2026
Assignee
International Business Machines Corporation
Inventors
Dennis Wei
Abstract
A computer-implemented method of decision-making using selective labels, includes receiving a conditional success probability value of a feature associated with an entity. A confidence value of the received success probability value is received. A parameter value that is a trade-off between a short-term learning and a long-term utility is selected. A decision is rendered to accept or reject the feature associated with the entity according to a machine learning policy.
CPC Classifications
G16H 50/20 G16H 20/13 G16H 40/67 G16H 50/70 G06N 5/045 G06N 20/00
Filing Date
2021-07-20
Application No.
17381141
Claims
17
Named provisions
Related changes
Source
Classification
Who this affects
Taxonomy
Browse Categories
Get Healthcare alerts
Weekly digest. AI-summarized, no noise.
Free. Unsubscribe anytime.
Get alerts for this source
We'll email you when ChangeBridge: Patent Grants - Health Informatics (G16H) publishes new changes.