⚛️ quantum physics

A physical and universal model of bosonic computations with Solovay-Kitaev theorem

This paper introduces Bosonic Energy-Preserving Quantum Computation (BEQC), a physically grounded model that restricts energy growth by using input coherent states and energy-preserving Hamiltonians, thereby establishing a universal and robust framework that recovers BQP, circumvents previous no-go results via a Solovay-Kitaev theorem, and enables high-precision state engineering and simulation.

Original authors: Dorian Rudolph, Arsalan Motamedi, Dhruva Sambrani, Hamid Reza Naeij, Ulysse Chabaud, Sevag Gharibian, Saeed Mehraban

Published 2026-10-01

📖 4 min read🧠 Deep dive

CC BY 4.0

Original authors: Dorian Rudolph, Arsalan Motamedi, Dhruva Sambrani, Hamid Reza Naeij, Ulysse Chabaud, Sevag Gharibian, Saeed Mehraban

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Quantum computing promises to solve problems that are currently impossible for classical machines, but the path to building these machines is fraught with physical constraints. One of the most promising approaches uses light, specifically particles of light called photons, to carry information. Unlike traditional computers that use bits that are either zero or one, these systems use continuous variables, such as the brightness or phase of a light wave, which can take on a vast range of values. This offers a natural way to process information with high precision and strong error correction. However, a fundamental problem has long plagued these light-based models: to perform complex calculations, the standard mathematical recipes often require the energy of the light to grow uncontrollably. In these theoretical models, the number of photons can explode to infinite levels, a scenario that is physically impossible to build in a laboratory and renders the mathematical tools used to design the computers useless.

A team of researchers has now proposed a new way to think about these light-based computers that respects the laws of physics. They introduced a model called Bosonic Energy-Preserving Quantum Computation. The core idea is simple but profound: treat energy not as something that can be created out of nothing during a calculation, but as a limited resource that must be supplied at the very beginning. In this new framework, the computer starts with a specific amount of energy stored in a single, bright pulse of light. The operations performed on this light are strictly designed so that they never change the total amount of energy in the system. If the light starts with a certain number of photons, it ends with exactly that same number, regardless of how complex the calculation becomes. By strictly enforcing this rule, the researchers eliminated the runaway energy growth that made previous models unphysical.

The researchers demonstrated that this new, physically grounded model is just as powerful as the best existing quantum computers. They proved that it can efficiently simulate any standard quantum computation that runs in polynomial time, a class of problems known as BQP. This means that by carefully managing the energy input, the system can perform the same complex tasks as qubit-based computers without ever violating physical laws. Furthermore, they showed that this model is robust. They identified specific sets of optical tools, such as beam splitters and special crystals that interact with light intensity, that can be combined to build any necessary calculation. Crucially, they proved that these tools can be assembled efficiently, a property known as the Solovay–Kitaev theorem, which ensures that the computer can be programmed with a manageable number of steps. This was a major hurdle in previous models, where the lack of a fixed set of tools made it difficult to guarantee that a calculation could be completed in a reasonable time.

Beyond proving the computer works, the team showed how to use it to prepare specific, highly useful states of light that are difficult to create otherwise. They described a method to generate complex patterns of light, known as Gottesman–Kitaev–Preskill states, which are essential for protecting quantum information from errors. They also showed how to create specific counts of photons with extreme precision. These results are not just theoretical; they provide a concrete blueprint for building a fault-tolerant quantum computer using light. The work also addressed a significant gap in the field by showing that this energy-preserving approach can simulate not only pure light-based systems but also hybrid systems that mix light with matter, as well as systems based on electrons.

The researchers were careful to rule out the idea that the old, unbounded models could simply be fixed by adding a promise that energy stays low. They showed that without the strict rule of energy-preserving operations, the mathematical tools used to design the circuits would fail, and the energy would still diverge in ways that make the computer unreliable. Their new model avoids this by making energy conservation a fundamental part of the design, rather than an afterthought. This approach ensures that the computer remains physically realizable while retaining the full computational power needed for universal quantum computing. The findings suggest that by viewing energy as a finite input resource, scientists can design quantum computers that are both mathematically sound and physically possible to build, bridging the gap between abstract theory and experimental reality.

Technical Summary: A Physical and Universal Model of Bosonic Computations with Solovay–Kitaev Theorem

1. Problem Statement

The paper addresses fundamental shortcomings in existing models of continuous-variable (CV) bosonic quantum computation, specifically the standard Lloyd–Braunstein (LB) model [LB99] and hybrid oscillator-qubit models. While these models are theoretically universal, they suffer from "unphysical" behavior due to unbounded energy growth.

The authors identify two critical issues with the LB model:

Unreasonable Computational Power: Standard gate sets (e.g., Gaussian gates plus a cubic phase gate $\hat{q}^3$ ) can generate energy (average photon number) that grows doubly exponentially with circuit depth or even diverges to infinity with constant time evolution. This allows the model to solve problems outside standard complexity classes (e.g., $NTIME(\exp(k))$ ), rendering it physically unrealistic.
Loss of Robustness: The unbounded energy prevents the existence of efficient, fixed-size universal gate sets. Specifically, the Solovay–Kitaev theorem (which guarantees efficient compilation of arbitrary unitaries from a finite gate set) fails for these models. Trotter splitting and commutator-based decompositions often fail to converge in the operator norm or require infinite gate sets to approximate physical unitaries within a desired precision.

The core challenge is to define a bosonic computational model that is physically grounded (energy remains bounded) while retaining the computational power and algorithmic robustness (universality, efficient compilation) necessary for a standard theory of computation.

2. Methodology: Bosonic Energy-Preserving Quantum Computation (BEQC)

To resolve these issues, the authors introduce Bosonic Energy-Preserving Quantum Computation (BEQC). The model is constructed on two primary principles:

Energy-Preserving Gates: All computational gates are generated by Hamiltonians that commute with the total photon number operator $\hat{N}$ (i.e., $[H, \hat{N}] = 0$ ). Consequently, these gates cannot change the total energy of the system. This restricts the dynamics to subspaces of fixed photon number.
Energy as an Input Resource: Since gates cannot generate energy, all energy required for computation must be supplied via the input state. The model utilizes a single coherent state $|\alpha\rangle$ (with amplitude $\alpha \in \mathbb{C}$ ) as a resource state, alongside vacuum modes. The total energy of the computation is strictly bounded by the input energy $|\alpha|^2$ .

This construction ensures that energy growth is impossible by design, making the model physically grounded. The authors formalize the complexity class BEQP (Bosonic Energy-Preserving Quantum Polynomial time), analogous to BQP, where computations run in polynomial time with polynomial energy bounds.

3. Key Contributions and Results

A. Computational Power and Universality

The paper establishes that BEQC is a robust and powerful computational model:

Equivalence to BQP: The authors prove that BEQP with polynomial energy and time is exactly equivalent to the standard discrete-variable class BQP (Theorem 7.1 and 7.2). This holds even if the input resource state is a displaced thermal state or a thermal state, demonstrating robustness against noise.
Complexity Hierarchy: By varying parameters (energy, precision, gate locality), the model captures a range of complexity classes. For instance, BEQP with non-local, polynomial-degree Hamiltonians is shown to be PSPACE-complete (Theorem 1.2), aligning with intuition that simulating non-sparse Hamiltonians or long-time evolutions can be harder than standard BQP tasks.
Simulation of Existing Models: BEQC can efficiently simulate the LB model, hybrid CV-DV models, fermionic computations (via Jordan-Wigner mapping), and fusion-based quantum computing, provided the target computations have polynomial energy moments (Theorem 1.3 and Section 5.3).

B. Universal Gate Sets and the Solovay–Kitaev Theorem

A major contribution is the proof of the existence of natural, finite universal gate sets for BEQC that satisfy a Solovay–Kitaev theorem (Theorem 1.4):

Gate Sets: The authors identify two universal sets based on linear optics (beam splitters and phase shifters) combined with either self-Kerr ( $e^{i\theta \hat{n}^2}$ ) or cross-Kerr ( $e^{i\theta \hat{n}_j \hat{n}_k}$ ) interactions. These gates are energy-preserving.
Efficient Compilation: They prove that any energy-preserving unitary $U$ acting on a subspace with total photon number at most $K$ can be approximated to precision $\epsilon$ using a circuit of size $\text{poly}(m, K, \log(1/\epsilon))$ . This circumvents previous no-go results that suggested such efficient compilation was impossible for bosonic systems.
Extension to Non-Energy-Preserving Gates: Using parametric approximation techniques, the authors show how to compile general unitaries (including those not energy-preserving) by utilizing the coherent state resource. This allows simulating active gates (like displacement or squeezing) with an error that scales inversely with the resource energy (Theorem 5.2).

C. State Synthesis and Applications

The framework enables deterministic state preparation without intermediate measurements or postselection:

Fock State Synthesis: The model can prepare arbitrary finite superpositions of Fock states with inverse-polynomial precision using a polynomial number of gates and resource energy (Theorem 1.6).
GKP State Synthesis: The authors provide a protocol for the deterministic preparation of high-quality Gottesman–Kitaev–Preskill (GKP) states, a crucial resource for bosonic error correction. The circuit uses linear optics and Kerr gates, with resource energy scaling polynomially with the desired precision (Theorem 1.7).
Bose–Hubbard Dynamics: The paper demonstrates the decomposition of the standard Bose–Hubbard Hamiltonian into energy-preserving gates, enabling efficient quantum simulation of these dynamics within BEQC (Appendix F).

D. Parametric Approximation

The authors rigorously analyze parametric approximation, a technique where non-energy-preserving active gates are simulated by energy-preserving gates acting on a high-energy coherent state ancilla. They prove that the simulation error scales as $O(1/\alpha)$ (where $\alpha$ is the coherent amplitude) and is robust against thermal noise in the resource state (Theorem 5.4).

4. Significance and Claims

The paper claims to provide the first "physically grounded" and "computationally robust" model of universal bosonic quantum computation. Its significance lies in:

Resolving the Energy Problem: By treating energy as a controlled input resource rather than an unbounded output of gate dynamics, BEQC avoids the "unphysical" energy blowups that plague the LB model.
Restoring Algorithmic Tools: The model recovers fundamental algorithmic tools lost in previous CV models, specifically the existence of fixed universal gate sets and the Solovay–Kitaev theorem for efficient compilation.
Bridging Theory and Practice: The model captures the power of BQP while remaining consistent with physical constraints (bounded energy), suggesting a viable path toward scalable, fault-tolerant bosonic quantum computing. The authors note that the energy control inherent in BEQC is a structural requirement for fault tolerance, as uncontrolled energy growth would populate arbitrarily high-energy sectors, undermining error correction assumptions.

The work does not propose specific experimental hardware but rather defines a theoretical framework that can guide the design of fault-tolerant architectures and clarify the computational limits of continuous-variable systems.