Project Description
Team of QuantumForGraphproblem has developed a new type of quantum algorithm for solving linear systems, which is an essential mathematical tool used across science, engineering, and machine learning. Unlike previous approaches, our algorithm leverages information about the specific problem instance to bypass a major bottleneck (the condition number) that limits existing quantum methods. As a result, it can run significantly faster, sometimes exponentially faster, on certain structured linear systems.
Leveraging this algorithm, we further designed an end-to-end framework for solving polynomial systems, a unifying formulation for numerous graph-theoretic and optimization problems. Our approach not only improves upon existing quantum linear system algorithms but also opens the door to achieving large quantum speedups in a wide range of applications where linear systems play a central role.
Core Innovation
A new quantum linear system algorithm that opens up possibilities for a wide range of applications with significant quantum advantage.
Phase I Submission to XPRIZE Quantum Application Competition
A New Quantum Linear System Algorithm Beyond the Condition Number and Its Applications (arxiv:2510.05588).
(Advanced to the finalist stage; Acceptance Rate: 7/133)
Quantum Applications (Tentative)
Combinatorial Optimization: such as Quadratic Unconstrained Binary Optimization (QUBO) and Satisfiability (SAT)
Graph problems: such as pathfinding, Maximum Independent Set.
Nonlinear differential equations
Ground state preparation
We are currently investigating the potential quantum advantage offered by our new algorithm in the above applications and examining how it can be linked to practical, real-world scenarios.
Current Team Members (Tentative; Alphabetically Sorted)