How to unitarily map between any two pure states with a single closed-form exponential

This paper presents a dimension-independent, basis-agnostic closed-form exponential unitary transformation that maps any two pure quantum states using a single generator, eliminating the need for explicit basis constructions required by previous methods.

The Gist

The Big Idea: A "Magic Button" for Quantum States

Imagine you are in a giant, multi-dimensional room (this is the Hilbert Space where quantum states live). You are standing at one specific spot (State A), and you want to get to another specific spot (State B).

In the quantum world, you can't just walk there; you have to "rotate" the entire room to move yourself from A to B without losing any information. This rotation is called a Unitary Transformation.

The Problem:
Until now, figuring out exactly how to rotate the room to get from A to B was like trying to navigate a maze by drawing a map of every single wall, floor, and ceiling first.

• The Old Way (Gram-Schmidt): To get from Point A to Point B, scientists used to have to build two complete, perfect grids (bases) of coordinates—one for where you are and one for where you want to go. Then, they would manually match every single point on Grid A to Grid B.
• The Flaw: This is messy, slow, and gets incredibly complicated as the room gets bigger (higher dimensions). It's like trying to solve a puzzle by building the whole box before you even look at the picture on the lid.

The New Solution (This Paper):
The authors, Peter, Marcus, and Jonte, found a shortcut. They discovered a way to write a single, elegant mathematical formula (a "closed-form exponential") that acts as a magic button. You press it, and it instantly rotates State A into State B.

You don't need to build the whole grid. You don't need to know the size of the room. You just need the two points, and the formula does the rest.

---

How It Works: The "Swing" Analogy

To understand their method, imagine two people, Alice and Bob, standing on a dance floor.

1. The Old Way: To get Alice to Bob's spot, you would first have to measure the entire dance floor, draw a coordinate system, find the exact angle for every single step, and then tell Alice to take a specific sequence of steps.
2. The New Way: The authors realized that Alice and Bob define a specific "plane" or "swing" between them.
   • They treat the space between Alice and Bob like a swing set.
   • Instead of mapping the whole room, they just calculate the perfect push needed to swing Alice directly to Bob.
   • This "push" is the Unitary Generator.

The Secret Sauce: The "Minimal Polynomial"

How did they find this perfect push without doing all the heavy math? They used a trick from algebra called the Minimal Polynomial.

Think of the "push" (the generator) as a machine with a few gears.

• Usually, machines have infinite gears, making them hard to predict.
• But the authors realized that the machine connecting two quantum states only has three gears (or fewer).
• Because the machine is so simple (it only has three gears), they could write a "cheat sheet" (the formula) that tells you exactly how the machine moves after any amount of time.

This allowed them to write the solution as a single exponential equation (like $e^{x}$), which is the mathematical equivalent of a smooth, continuous rotation.

Why Does This Matter? (The "Why Should I Care?")

1. It's Faster and Cleaner
In quantum computing, we need to prepare specific states to run algorithms. The old method was like trying to build a house by laying every single brick individually. The new method is like using a 3D printer to print the whole wall in one go. It's much more efficient.

2. It's "Dimension Agnostic"
The old method broke down if the room was huge (like a million dimensions). The new method works just as well in a tiny room as it does in a massive one. It doesn't care how big the universe is; it only cares about the relationship between the two points.

3. It Reveals the "Rotation"
The authors point out that a unitary transformation is essentially a rotation in complex space.

• If you are in a real world (like a 3D room), a rotation is simple.
• In the quantum world, there is also a "phase" (like a hidden color or timing) attached to the rotation.
• Their formula shows exactly how to handle this hidden phase. It's like realizing that to get from A to B, you don't just turn left; you turn left and spin your hat at the same time. Their formula calculates both moves perfectly.

Summary in a Nutshell

• The Goal: Move a quantum particle from State A to State B.
• The Old Way: Build a massive map of the whole universe, then draw a line. (Slow, messy, requires knowing the size of the universe).
• The New Way: Use a single, smart formula that calculates the direct "swing" between A and B. (Fast, clean, works for any size universe).
• The Result: A "magic button" (a closed-form exponential) that quantum engineers can use to program quantum computers more efficiently, without getting bogged down in complex math.

The paper essentially says: "Stop building the whole map. Just look at the two points, find the swing between them, and push."

Technical Summary

1. Problem Statement

In quantum information theory, it is a fundamental fact that any two pure quantum states within the same Hilbert space can be mapped to one another via a unitary transformation. However, constructing this specific unitary operator is non-trivial.

• Current Limitations: Traditional approaches (e.g., using the Gram-Schmidt process) require:
  1. Constructing two complete orthonormal bases (one containing the initial state, one containing the target state).
  2. Defining a pairwise mapping between these bases.
  3. The complexity of this method scales with the dimension of the Hilbert space ($d$), making it computationally "messy" and inefficient for high-dimensional systems.
  4. These methods often fail to express the transformation as a single exponential of a Hermitian operator (a "rotation"), obscuring the geometric intuition of unitary evolution.
• Goal: The authors aim to derive a single, closed-form exponential unitary operator that maps an arbitrary initial pure state $|\psi_0\rangle$ to a target pure state $|\phi_0\rangle$. This method must be:
  • Independent of any specific basis choice.
  • Agnostic to the dimension of the Hilbert space.
  • Constructed using a single generator.

2. Methodology

The authors employ an abstract algebraic approach rather than matrix-based linear algebra. The core of their methodology involves the following steps:

A. Abstract Hilbert Space Definition

The work is framed in an abstract complex Hilbert space $\mathcal{H}$ defined by a vector space $V$ and a sesquilinear form (inner product) $h$. This allows the derivation to remain valid for any dimension without referencing specific coordinates.

B. Unitary Generators via the Map $t(a,b)$

Instead of using standard matrix generators, the authors define a specific map $t: V \times V \to (V \to V)$:
$$t(a,b)(c) = h(a,c)b - h(b,c)a$$

• This map is shown to be anti-Hermitian (the generator of unitary transformations).
• The set of such maps forms a real Lie algebra isomorphic to $\mathfrak{u}(n, \mathbb{C})$.
• Crucially, the action of $t(a,b)$ on any vector is confined to the subspace spanned by $\{a, b, c\}$, effectively reducing the problem to a low-dimensional subspace regardless of the total Hilbert space dimension.

C. Minimal Polynomial Analysis

The authors derive the minimal polynomial of the operator $t(a,b)$. By analyzing the algebraic properties of $t(a,b)$, they determine that its minimal polynomial is at most cubic. They define real quantities based on the inner products of the states $a$ and $b$:

• $g = \frac{1}{2}(h(b,a) + h(a,b))$ (Real part)
• $\omega = \frac{1}{2i}(h(b,a) - h(a,b))$ (Imaginary part)
• $H^2 = h(a,a)h(b,b) - h(a,b)h(b,a)$ (Related to the Cauchy-Schwarz inequality)
• $G^2 = H^2 + \omega^2$

D. Case Analysis and Projection

The derivation splits into two cases based on the value of $H$:

1. Case $H \neq 0$: The states are linearly independent over $\mathbb{C}$. The minimal polynomial is cubic. The authors construct projection operators ($\Pi_0, \Pi_1, \Pi_2$) onto the eigenspaces of $t(a,b)$ purely algebraically, without needing to find eigenvectors.
2. Case $H = 0$ (but $t(a,b) \neq 0$): This occurs when $b = \alpha a$ (linear dependence over $\mathbb{C}$). The minimal polynomial reduces to quadratic, and different projection operators are derived.

E. Closed-Form Exponential Construction

Using the resolution of identity provided by the projection operators, the authors compute the matrix exponential $\exp(\theta t(a,b))$ in closed form. They solve for the specific angle $\theta$ required to map state $a$ to state $b$ (up to a global phase).

3. Key Results

The Closed-Form Solution

The paper provides explicit formulas for the unitary operator $U = \exp(\theta t(a,b))$ that maps $|a\rangle$ to $|b\rangle$:

• For $H \neq 0$:
  The exponential is given by a complex expression involving trigonometric functions of $\theta$ and the parameters $g, \omega, G, H$.
  $$\exp\left(\frac{\theta t(a,b)}{G}\right) = \text{Id} + \frac{t(a,b)}{H^2}\{\dots\} + \frac{t(a,b)^2}{H^2}\{\dots\}$$
  The required angle $\theta'$ satisfies $\cot(\theta') = g/G$. The resulting state is proportional to $b$:
  $$\exp\left(\frac{\theta' t(a,b)}{G}\right) |a\rangle \propto |b\rangle$$

• For $H = 0$:
  The formula simplifies significantly:
  $$\exp\left(\frac{\theta t(a,b)}{\omega}\right) = \text{Id} + \frac{t(a,b)}{\omega} \{ e^{-i\theta} \sin(\theta) \}$$
  Here, the transformation primarily adjusts the phase of the state.

Connection to Rodrigues' Rotation Formula

The authors demonstrate that when the imaginary part of the inner product vanishes ($\omega = 0$), the derived formula reduces exactly to the famous Rodrigues' rotation formula for 3D rotations. This highlights that unitary transformations in complex Hilbert space are a generalization of orthogonal rotations, with the phase of the inner product introducing interference effects.

4. Significance and Applications

• Basis and Dimension Independence: The method does not require constructing full bases or knowing the dimension of the Hilbert space. It relies solely on the algebraic relationship between the two specific states involved.
• Quantum State Preparation: The closed-form exponential is highly valuable for quantum computing. Current state preparation often involves "dialing in" amplitudes via binary trees and Gray codes. A single exponential generator could potentially be decomposed more efficiently into elementary gates (shuffles and rotations) than the brute-force Gram-Schmidt approach.
• Geometric Intuition: By expressing the map as a single exponential of a generator, the work reinforces the view of unitary evolution as a "rotation" in Hilbert space, bridging the gap between abstract algebra and geometric intuition.
• Future Directions: The authors suggest this algebraic framework could be extended to:
  • Approximating operators (relaxing the exactness condition).
  • Transformations between mixed states.
  • Optimizing quantum circuits by identifying the most direct algebraic path between states.

Conclusion

Bradshaw, Gouveia, and Hance present a rigorous algebraic solution to the problem of mapping pure quantum states. By utilizing the minimal polynomial of a specific anti-Hermitian generator $t(a,b)$, they derive a single, closed-form exponential unitary that transforms any pure state to any other. This approach eliminates the need for basis construction, scales independently of Hilbert space dimension, and offers a powerful new tool for quantum circuit design and state preparation.