Researchers built a quantum device that produces a superposition of all the possible futures, a technology that can help futuristic artificial intelligence learn much faster than it can today.
The team from Griffith University and Nanyang Technological University implemented a specially devised photonic quantum information processor in which the potential future outcomes of a decision process are represented by the locations of the photons.
They then showed that the state of the quantum device was a superposition of multiple potential futures, weighted by their probability of occurrence.
At present, the quantum computer created can hold two superpositions of 16 different possibilities simultaneously. The quantum device also uses less memory than a classical computer would, which means that it can outperform the classical systems at certain tasks too.
Currently, artificial intelligence learns by analyzing example after example and by looking at different patterns. The team behind the quantum computer believes that their quantum superpositions can greatly enhance this AI learning process.
“Our approach is to synthesize a quantum superposition of all possible futures for each bias,” said scientist Farzad Ghafari. “By interfering these superpositions with each other, we can completely avoid looking at each possible future individually.”