Deep learning with judgment

This system enhances a deep learning network by adding a special "judgment" component. This component observes how another part of the network performs on individual data items and decides if its output is "correct." Uniquely, while this judgment influences the overall network training, it specifically prevents its feedback from directly changing the observed part of the network.

Why it matters: Filed as large AI models gained prominence, the need for more controlled and interpretable training methods has intensified since 2023. This selective feedback approach could be valuable for integrating expert knowledge or specific objectives into complex, opaque models without disrupting their core learning.

Status: Not in forceListed as no longer active. The specific reason is not in the record we hold.
How hard to build: SpecializedSoftware logic, neural network architecture, training algorithms

3 AI reinvention directions for this patent

See all three

AI gives you a few directions you could take this. Pick one, and we check whether your version is different enough to patent, then write the filing.

Reinvent this with AI

Where this one came from

Every morning we go through millions of abandoned patents and pick out 15 worth your time. This was one of today’s.

This one

+10