Researchers from the University of Michigan, led by physicist Enrico Rinaldi, have utilized quantum computing and machine learning to gain fresh insights into the nature of black holes. This groundbreaking study, published on February 3, 2025, in PRX Quantum, explores the quantum state of matrix models, advancing our understanding of black hole physics.
The research is grounded in the holographic principle, suggesting a mathematical equivalence between fundamental theories of particle physics and gravity, despite their formulation in different dimensions. Two prevailing theories describe black holes through distinct dimensional frameworks, with gravity operating in three-dimensional space while particle physics is confined to a two-dimensional surface.
This duality emphasizes the interconnected nature of both models, as the immense mass of a black hole distorts spacetime, creating a gravitational field extending into three dimensions. This gravitational influence mathematically correlates with particles moving in two dimensions above the black hole. Some scientists propose that the entire universe may function similarly, as a holographic projection of particles.
Rinaldi and his team investigated how quantum computing and deep learning can enhance research on holographic duality. Their focus was on calculating the ground state energy of quantum matrix models, which may unravel the underlying nature of this duality. These models represent particle theory, where mathematical events in one system can affect another representing gravity.
By solving relatively simple matrix models that encapsulate the characteristics of more complex models used to describe black holes, the researchers aim to understand particle theory properties, potentially yielding insights into gravity. Rinaldi noted, “Understanding the properties of this particle theory through numerical experiments may reveal something about gravity.”
The study employs quantum circuits, represented as wires connected to qubits—quantum information bits—where quantum operations dictate the flow of information. Rinaldi likened the process to music, where each step transforms qubits into new forms, ultimately achieving the ground state.
Through their work, the researchers successfully identified the ground state of two matrix models using both quantum circuits and traditional methods, despite current hardware limitations on qubit numbers. Rinaldi emphasized that while conventional methods can find ground state energy, they often fail to provide the complete wave function structure.
These findings represent a critical step towards future research on quantum algorithms and machine learning applications in exploring quantum gravity via holographic duality. Rinaldi’s team plans to extend their results to broader matrices and assess their resilience against noise effects that could introduce inaccuracies.