Quantum AI Breakthrough: Theorem Shrinks Appetite for Training Data

Less data for more results. 

Quantum AI Breakthrough: Theorem Shrinks Appetite for Training Data

Los Alamos National Laboratory

August 23, 2022

A proof devised by a multi-institutional team of scientists demonstrates that quantum neural networks can train on minimal data. "The need for large datasets could have been a roadblock to quantum AI [artificial intelligence], but our work removes this roadblock," said Patrick Coles at the U.S. Department of Energy's Los Alamos National Laboratory (LANL). Coles said quantum AI training occurs in a mathematical construct called a Hilbert space, and the theorem shows that navigating this space requires only as many data points as the number of parameters in a given model. The researchers could ensure that a quantum model can be compiled in far fewer computational gates relative to the volume of data. LANL's Marco Cerezo said, "We can compile certain very large quantum operations within minutes with very few training points—something that was not previously possible." ... '