Invention:
This invention is a method using quantum entanglement to increase the performance of machine learning. Here, supervised learning is enhanced by an entangled sensor network. This method allows for supervised learning tasks at the physical layer where a quantum advantage is achieved. The entanglement is shared by different sensors which boost the performance of extracting the global features of the object under investigation. This enabling technology offers the potential for ultrasensitive measurements in biological, thermal, and mechanical systems.
Background:
Current quantum machine learning algorithms aim to efficiently solve certain tasks that are intractable by the optimum classical algorithm; however, achieving the quantum advantage is held back by various challenges.
Applications:
- Quantum supervised learning
- Ultra-sensitive measurement
Advantages:
- Performance gain in extracting features of object under investigation
- Utilizes off-the-shelf components
- Enables a near-term, practically useful device
Status: issued U.S. patent #12,067,456
