Skip to content

Results are limited to the current section: Application solving tools

Product news

Qubo Solver

Qubo Solver is a Python library for solving Quadratic Unconstrained Binary Optimization (QUBO) problems on Pasqal's neutral-atom quantum processors.

A QUBO instance is defined by a symmetric matrix QQ of size nร—nn \times n. Solving it means finding the binary vector xx of length nn below โ€” a form general enough to express many combinatorial optimization problems, from finance to machine learning.

minโกxโˆˆ{0,1}nโ€…โ€ŠxโŠคQx=โˆ‘i,jqijxixj\min_{x \in \{0,1\}^n} \; x^\top Q x = \sum_{i,j} q_{ij} x_i x_j

Qubo Solver implements algorithms to solve QUBO instances: quantum, classical, and hybrid quantum-classical.

To be solved on a neutral-atom quantum processor, a QUBO is embedded onto a neutral-atom register, driven with a laser pulse designed to steer the atoms toward low-cost bitstrings, and the resulting program is executed and sampled on a QPU or emulator, all behind a simple and flexible interface.

Q=(q11โ‹ฏq1nโ‹ฎโ‹ฑโ‹ฎqn1โ‹ฏqnn)Q = \begin{pmatrix} q_{11} & \cdots & q_{1n} \\ \vdots & \ddots & \vdots \\ q_{n1} & \cdots & q_{nn} \end{pmatrix}
โญฉ            โญจ

Register

Drive
โญจ            โญฉ
   labels       bitstrings       costs  counts  probs
0       0  100101100000010 -205.649830       1  0.001
1       0  100000100000010 -199.302937       2  0.002
2       0  100000110000010 -178.590227       1  0.001
3       0  100001100000000 -177.514687       2  0.002
4       0  100000100001000 -175.325455       2  0.002
5       0  100000000001010 -169.498798       3  0.003
6       0  100100100000010 -167.246616       1  0.001
9       0  100000010001010 -148.786089       1  0.001
10      0  100100110000010 -146.533907       1  0.001
11      0  100101100000000 -145.458366       2  0.002
12      0  100000110000000 -140.964319       5  0.005
13      0  100101000000010 -139.631710       2  0.002
14      0  100000100000000 -139.111473      17  0.017

With Qubo Solver you can:

  • define QUBO instances from symmetric matrices,
  • solve them with classical, quantum, or hybrid solvers, on a QPU or an emulator,
  • pre- and post-process instances and solutions to improve solving quality.

Qubo Solver is designed for both newcomers to quantum computing, who can implement and run a solver in a few simple steps, and experienced users, who will find a flexible environment for exploring new solving methods while interacting with a simple, intuitive QPU interface.