Uncertainty-aware Feedback Loop Diagnosis (UFLD)
What it is
UFLD is an algorithm for wafer defect classification that uses reinforcement learning to adaptively decide when to stop acquiring measurement data. It integrates conformal prediction to provide uncertainty estimates, allowing the system to stop once sufficient information is gathered, which reduces scan time and improves efficiency.
The problem it solves
Existing deep learning approaches for wafer defect detection require a full surface scan, which hinders the development of high-throughput inspection systems and increases operating costs.
What you could use it for
- High-throughput semiconductor wafer inspection
- Reducing scan time in defect detection systems
- Reliable production monitoring in multi-step manufacturing
Why it matters
Significantly reduces inspection time (by ~40%) while maintaining high accuracy (98.4%), enabling faster and more cost-effective high-volume semiconductor manufacturing.
Read the source ↗Back to the library →
Summary written by an AI model from the source; check the source before you rely on it.