Combined data pre-process and architecture search for deep learning models

This invention automatically finds the best combination of an AI model and the way data should be prepared for it, tailored to a specific task. It does this by training a recurrent neural network to rank various combinations of AI models and data preparation strategies, selecting the top-performing one.

Why it matters: Filed when automated deep learning optimization was computationally intensive. By 2026, advancements in AutoML frameworks and cloud computing have made such complex, multi-objective searches significantly more feasible and efficient for diverse deep learning tasks.

Status: Not in forceListed as no longer active. The specific reason is not in the record we hold.
How hard to build: SpecializedSoftware logic, deep learning expertise

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