Unitary Dilation Technique

• Unitary dilation technique is an operator-theoretic method that embeds non-unitary (usually contraction) operators as compressions of unitary ones in an extended Hilbert space.
• It employs canonical constructions via defect operators and block matrices, generalizing from the Sz.-Nagy theorem to continuous-variable and discrete quantum simulation frameworks.
• This approach underpins diverse applications including quantum simulation and realization theory, offering scalable algorithms with controlled precision and resource efficiency.

A unitary dilation technique is any operator-theoretic procedure for embedding a prescribed non-unitary operator or semigroup (typically, a contraction or a contraction semigroup) acting on a Hilbert space $\mathcal H$ as the compression of a unitary or normal operator or evolution on a larger Hilbert space $\mathcal K \supseteq \mathcal H$. Unitary dilation provides an exact or approximate representation:

$$T^n = P_\mathcal{H}\,U^n|_\mathcal{H} \qquad\text{or} \qquad V(t) = P_\mathcal{H} U(t) |_\mathcal{H},$$

where $P_\mathcal{H}$ is the orthogonal projection and $U$ or $U(t)$ is unitary (or a one-parameter unitary group), with minimality and uniqueness properties often enforced. Dilation theory underpins significant developments in operator theory, function theory, and, increasingly, quantum simulation.

1. The Classical Unitary Dilation Theorem and Its Canonical Construction

The Sz.-Nagy unitary dilation theorem asserts that every contraction $T \in B(H)$ (i.e., $\|T\| \le 1$) on a Hilbert space $H$ admits a unitary dilation: there exists a Hilbert space $K \supseteq H$ and a unitary $U \in B(K)$ such that

$T^n = P_H U^n |_H, \qquad \forall n \ge 0$

where $P_H$ is the orthogonal projection onto $H$. $K$ can be chosen minimally so that $$K = \overline{\mathrm{span}} \{ U^n h : n \in \mathbb Z, h \in H \}$$, and under this choice the pair $(U, K)$ is unique up to unitary equivalence fixing $H$ (Shalit, 2020).

The canonical construction involves identifying the defect operators $D_T = (I - T^* T)^{1/2}$, $D_{T^*} = (I - T T^*)^{1/2}$, and the corresponding defect spaces $\mathcal H_+ = \overline{\mathrm{Ran}}\,D_T$, $\mathcal H_- = \overline{\mathrm{Ran}}\,D_{T^*}$. The minimal unitary dilation acts on $K = H \oplus \mathcal H_+ \oplus \mathcal H_-$ and is given by the block matrix

$$U = \begin{bmatrix} T & D_{T^*} & 0 \ D_T & -T^* & 0 \ 0 & 0 & I_{\mathcal H_-} \end{bmatrix}.$$

This construction generalizes via Wold-type decompositions and admits extensions to tuples of commuting contractions, spectral set contexts, and beyond (Shalit, 2020, Pal et al., 2022).

2. Schrödingerisation and Continuous-Variable/Discrete Dilation

The "Schrödingerisation" unitary dilation applies to contraction semigroups $V(t) = e^{-A t}$ with $A$ a bounded operator whose spectrum satisfies $\mathrm{Re}\,\lambda_j(A) \ge \lambda_0 > 0$. There exists an enlarged space $\mathcal H_1 \supseteq \mathcal H$ and a unitary group $U(t)$ so that $V(t) = P_{\mathcal H} U(t)|_{\mathcal H}$ for all $t \ge 0$. The construction is explicit both in continuous-variable (CV) and finite-dimensional (qubit) forms (Hu et al., 2023):

CV Construction: Decompose $A = H_1 + i H_2$ (Hermitian parts). Introduce an additional continuous variable $\eta \in \mathbb R$ and define

$$U_{CV}(t): h(\eta) \mapsto e^{-i(\eta H_1 + H_2)t} h(\eta).$$

The embedding and projection are specified, and the action satisfies

$$P_{\mathcal H} U_{CV}(t) h = e^{-H_1 |t| - i H_2 t} h = V(t) h.$$

Discrete/Qubit Version: Discretize $\eta$ to $N$ points, forming the matrix $D = \mathrm{Diag}(\eta_0, \ldots, \eta_{N-1})$ and the Hamiltonian

$H^{(DV)}_{Schr} = H_1 \otimes D + H_2 \otimes I_N.$

The unitary $U_{DV}(t) = e^{-i H^{(DV)}_{Schr} t}$ approximates $V(t)$ to error $O(\delta)$ for $L = O(1/\delta)$, $N = O(\tau/\delta^2)$ with $\tau = \|A\| t$, using a modest number of ancilla qubits (Hu et al., 2023).

3. Quantum Simulation and Complexity Implications

For quantum simulation, the dilation technique enables simulation of nonunitary dissipative dynamics via embedding into a unitary evolution on an extended Hilbert space. In the Schrödingerisation framework, continuous-variable simulators (e.g., trapped ions, optical modes) can implement the exact dilation using one additional qumode. Digital quantum computers discretize the dilation, requiring $O(\log 1/\delta)$ ancilla qubits and $O(\delta^{-3})$ gate complexity for precision $\delta$.

Compared to block-encoding + QSVT methods, Schrödingerisation is independent of the spectral gap $\lambda_0$ (which may diverge as mesh size $h \to 0$ in PDEs), though its precision dependence may be polynomially worse. For prototypical applications (e.g., the discretized heat equation), the two schemes yield similar scaling in spatial grid size and evolution time (Hu et al., 2023).

4. Connection to Realization Theory and Interpolation

Unitary dilation techniques unify operator-theoretic realization frameworks. The dilation-theoretic approach yields the state-space realization or transfer-function model of Schur-class (contractive) functions on the disk, expressing $f(z)=D+zC(I-zA)^{-1}B$, with $T=[A\,B;C\,D]$ contractive. Rational inner approximants arise via transfer-functions of finite-dimensional unitary dilations (colligations), converging to arbitrary Schur functions and their matrix-valued analogues (Alpay et al., 2022). Potapov and Kreĭn–Langer factorizations further connect thereby to approximation of functions with prescribed kernel negative squares, and $J$-contractive settings.

5. Algorithmic/Quantum Implementations

Emerging quantum computing methods leverage the explicit structure of unitary dilation. Approaches include:

• Block-encoding and LCU: Embedding non-unitary operators via block encoding and linear-combination-of-unitaries for efficient quantum circuits (Hu et al., 2023, Mehta et al., 30 Jan 2025).
• Single-ancilla stochastic protocols: Exact decomposition of any Kraus operator as a finite linear combination of unitaries, with a single ancilla qubit controlling all cross-terms. Lagrange-Sylvester interpolation ensures no approximation error. This dramatically reduces the number of measurements compared to finite-difference-based schemes (Mehta et al., 30 Jan 2025).
• Variational Unitary Dilation: NISQ-oriented hybrid algorithms embedding a nonunitary operator as a principal block of a parameterized unitary circuit, with cost functions such as Choi-state fidelity minimizing the residual (Li et al., 22 Oct 2025).

These schemes realize highly resource-efficient simulation of general open-system dynamics, including strong (noncontractive) and non-Markovian processes, with observed gate-count and measurement reductions of orders of magnitude on experimental quantum hardware (Li et al., 22 Oct 2025, Mehta et al., 30 Jan 2025).

6. Comparisons, Limitations, and Broader Context

The unitary dilation technique, in its various forms, is universally applicable to contraction operators and contraction semigroups; minimality and uniqueness (up to unitary equivalence) hold in the single-operator case. For multivariable, freely-independent, or noncommutative settings, the existence and structure of commuting or -freely-independent unitary dilations is more nuanced and often requires C-algebraic or Stinespring-dilation machinery (Atkinson et al., 2016, Pal et al., 2022).

Operationally, exact dilations for noncontractive operators are generally unattainable with classical Sz.-Nagy-like approaches; alternative constructions using biorthogonal or moment-matching representations extend the dilation toolbox for such settings (Koukoutsis et al., 2024, Li, 14 Jul 2025).

In quantum simulation, dilation-based approaches provide not only a physical pathway for simulating nonunitary processes on unitary hardware but also the theoretical foundation for analyzing the circuit complexity of quantum channels relative to resource penalties and environmental overhead, as formalized in recent Riemannian geometric frameworks (Acevedo et al., 2 Jan 2026).

7. Summary Table: Key Regimes and Dilation Techniques

Context	Dilation Mechanism	Key Features
Operator contraction	Sz.-Nagy unitary dilation	Unique, minimal, block-matrix explicit
Semigroup $e^{-At}$	Schrödingerisation, block-encoding	CV/qubit variants, analog/digital
Quantum open systems	LCU, stochastic, variational	Ancilla-efficient, NISQ-robust
Approximation/realization	State-space transfer, inner approx	Potapov/Kreĭn-Langer factorization
Multivariable/free probability	C*-product/Stinespring dilation	*-freeness, interplay with operator algebras

Unitary dilation remains the central paradigm for relating non-unitary evolutions to unitary dynamics in both operator theory and quantum simulation. Techniques continue to proliferate, adapting to high-dimensional, non-Hermitian, multivariable, and resource-constrained regimes, with explicit error bounds, complexity estimates, and quantum-circuit realizations now central to the field (Hu et al., 2023, Mehta et al., 30 Jan 2025, Li et al., 22 Oct 2025, Park, 20 Sep 2025).