QUBO (Quadratic Unconstrained Binary Optimization) shows up a lot in quantum optimization, but the core idea is actually pretty simple once you map it from a familiar problem.

Take MaxCut on a weighted graph:

You assign each node a binary variable (0 or 1)

The goal is to maximize the weight of edges that cross between the two partitions

You can rewrite this as a quadratic objective over binary variables:

Each edge contributes a term depending on whether its endpoints differ

Expanding this gives you a quadratic form: xᵀQx

The interesting part is:

The graph structure gets encoded directly into the Q matrix

Optimization becomes minimizing (or maximizing) a quadratic function over {0,1} variables

Example (2-node edge with weight w):

Contribution becomes something like: w(x₁ + x₂ - 2x₁x₂)

Scaling this up builds the full QUBO matrix.

What I find useful is thinking of QUBO as: → “just a way to turn combinatorial structure into a quadratic energy function”

This perspective makes it much less “quantum” and more like classical optimization with a different representation.

If anyone’s worked with QUBO in practice (classical or quantum solvers), I’d be curious:

Do you think in terms of matrices, or constraints first?

(There’s also a full walkthrough with a notebook demo here if you want more detail: https://youtu.be/P9sM2M-ahvs )

[留言]

为什么值得关注

原内容本身有足够细节，不是表面信息；符合当前抓取需求；原内容本身有足够细节，不是标题党或空洞总结

来源：reddit，领域：projects，保留分：0.55