Scientists at the Flatiron Institute’s Center for Computational Quantum Physics, working with collaborators at Boston University, demonstrated that classical computers can solve certain quantum dynamics problems using sophisticated mathematical approaches and specialized software. The research, published in Science, challenged claims made in an earlier study that asserted such calculations were beyond the capabilities of traditional computing hardware.
The specific challenge involved simulating the behavior of hundreds of interacting qubits arranged in various lattice configurations. Qubits, the quantum equivalents of classical bits, can exist in multiple states simultaneously, a property that makes modeling their interactions on conventional computers extraordinarily difficult. A particular obstacle was quantum entanglement, wherein connected qubits cannot be modeled independently despite physical separation, requiring complex algorithms to describe entire systems accurately. The wave function describing such systems grows exponentially in size with each additional particle, creating severe computational storage and processing constraints.
The breakthrough came through application of tensor network mathematics, which compresses wave function data into manageable mathematical structures. Researcher Joseph Tindall compared the approach to a compression algorithm that consolidates extensive information into interconnected numerical tables. Using the ITensor software library developed at their institution, researchers successfully performed simulations on standard laptop computers. The team also adapted belief propagation, an algorithm from the 1980s, for quantum applications, achieving state-of-the-art accuracy with modest computational resources while producing results matching previous quantum computer calculations.
The researchers emphasized that classical and quantum computing are complementary rather than purely competitive fields. Classical simulations can help validate quantum computer capabilities while simultaneously guiding hardware development. The findings suggest that optimized classical approaches may expand the range of quantum problems scientists can investigate without requiring quantum hardware access.
Looking forward, the team aims to extend these methods beyond qubit-only systems to model electrons moving between different sites, a considerably more complex simulation challenge with direct applications to understanding real quantum materials and superconductors.