Continuous variable quantum computing

Welcome! Before we map out our journey into continuous-variable quantum computing, I want to make sure this session is perfectly tailored to how your mind works.

Given your background in analog AI and memristors, this is going to be a fascinating bridge. To help me set the right pace and focus, which of these sounds most appealing to you?

A balanced mix of both math and hardware.

A balanced path is the perfect way to tackle this, especially since continuous-variable quantum computing bridges the gap between physical wave mechanics and pure mathematics.

Your path

Resources and Gaussian Operations

Achieving Universality: Non-Gaussianity

Error Correction and GKP Encoding

Applications and Future Perspectives

The Digital Qubit vs. The Analog Qumode

In standard quantum computing, we work with qubits. A qubit is inherently digital: it is a two-level system whose state lives in a finite, two-dimensional spanned by the states

|0\rangle

and

|1\rangle

. You can think of it like an electronic flip-flop that can exist in a superposition of being on and off.

Continuous-variable (CV) quantum computing flips this digital paradigm on its head. Instead of qubits, it uses qumodes (quantum modes). A qumode is the quantum equivalent of a harmonic oscillator—like a single swinging pendulum or a specific frequency of light. Rather than having only two states, a qumode can theoretically hold an infinite number of states because it operates in an infinite-dimensional Hilbert space. This makes the CV paradigm "analog" in spirit.

A discrete quantum bit shown on a Bloch sphere next to a continuous quantum mode shown as a smooth wave.

We have set up our map for this journey, starting with these continuous-variable foundations.

How does an infinite Hilbert space help with computing?

That is the million-dollar question, and to answer it, we can look at the sheer information capacity that comes with infinity.

In a digital quantum computer, a single qubit holds a state that is a combination of just two basic vectors,

|0\rangle

and

|1\rangle

. If you want to store more complex information, you have to string many qubits together, which requires entangling them and managing a lot of physical hardware. With a qumode, the physical state is a continuous wave. Because this wave lives in an infinite-dimensional space, a can theoretically store and process an entire continuous probability distribution all by itself. It is like the difference between trying to draw a smooth curve using a series of individual pixels versus using a single, continuous brushstroke.

This continuous nature makes CVQC exceptionally well-suited for simulating real-world physics. Most physical systems in chemistry, quantum field theory, and mechanics are fundamentally continuous—like the vibration of a molecule or the spreading of an electromagnetic wave. Instead of translating these smooth, analog processes into millions of binary 0s and 1s, we can map them directly onto the natural, continuous states of our qumodes.

A stepped digital binary bar graph trying to approximate a wave next to a smooth continuous wave perfectly matching the envelope.

I have created a quick visualization to show how this works. You can see how trying to force a smooth, analog wave into a digital qubit setup introduces blocky approximation steps, whereas the qumode handles the continuous wave naturally.

To see how we actually measure these continuous states in the real world, we use continuous physical observables. Would you like to explore how we define and measure these observables using physical waves?

Create a study guide on CV vs DV paradigms

I have put together a comprehensive study guide comparing the discrete-variable (DV) and continuous-variable (CV) paradigms to help you keep these fundamental concepts clear.

- Discrete-variable quantum computing relies on qubits that operate within a finite two-dimensional [{<mathematical space>}]. - Continuous-variable systems use

- **Q:** What is the fundamental unit of information in continuous-variable computing?<br> **A:** The qumode. - **Q:** How do the state spaces of qubits and qum

- **Qubit:** A discrete quantum bit that exists in a two-level state system. - **Qumode:** A continuous quantum mode capable of holding infinite states. - **Hil

DV vs CV QuantumStudy Guide

The study guide is ready for you to dive into. Now that we have contrasted these two main ways of thinking about quantum information, we can look under the hood at how we actually track and measure the physical state of a wave using continuous variables.

How do you physically measure a continuous wave?

Physically measuring a continuous quantum wave is all about looking at its fundamental properties: how tall the wave is (its amplitude) and where it is in its cycle (its phase). In continuous-variable quantum computing, we map these two features to physical mathematical observables called field quadrature operators, usually written as the position-like quadrature

\hat{x}

and the momentum-like quadrature

\hat{p}

These physical quadratures act exactly like the position and momentum of a quantum harmonic oscillator. Because they are quantum observables, they do not commute. They obey the famous which leads directly to the Heisenberg uncertainty principle.

[\hat{x}, \hat{p}] = i\hbar

To actually measure these values in a photonic lab, we use a beautiful optical technique called homodyne detection. Instead of trying to detect single photons directly, we take our weak quantum signal and mix it with a highly stable, much stronger reference laser called a on a beam splitter.

A optical homodyne detection setup where a weak quantum signal mode and a strong local oscillator mix at a beam splitter before detection.

I have drafted a visualization showing the setup of a homodyne detector. By adjusting the phase of that strong reference laser, we can selectively choose to measure only the amplitude

\hat{x}

, only the phase

\hat{p}

, or any linear combination of the two.

explain how does homodyne detection acutally work

To understand how homodyne detection works physically, we have to look at the wave interference that happens when we combine our delicate quantum signal with that powerful reference laser.

The Physics of Beam Splitting and Amplification

At the heart of a homodyne detector is a 50:50 —a simple semi-reflective mirror. We send two light beams into it: the weak quantum signal we want to measure, and a classical, highly stable reference laser of the exact same frequency, known as the local oscillator (LO). Since the local oscillator is incredibly strong compared to our fragile quantum signal, its classical wave amplitude acts like an physical amplifier.

When these two waves overlap on the beam splitter, they interfere. Depending on their relative phase, they will either add together (constructive interference) or subtract from each other (destructive interference). The beam splitter outputs two mixed beams, which are sent directly to two high-speed photodetectors. By subtracting the electrical currents from these two detectors, we cancel out the massive, noisy background of the local oscillator, leaving behind a clean signal directly proportional to the quantum wave's quadrature amplitudes.

\hat{I}_{diff} \propto E_{LO} \left( \hat{x} \cos\theta + \hat{p} \sin\theta \right)

This is where the magic happens. By physically shifting the phase $\theta$ of the local oscillator, we can choose exactly what we measure. If we set the phase to $\theta = 0$ , the sine term vanishes and we measure the amplitude quadrature $\hat{x}$ . If we shift the phase by ninety degrees to $\theta = \pi / 2$ , the cosine term vanishes and we measure the phase quadrature $\hat{p}$ .