Quantum Resource Quantifiers

• Quantum resource quantifiers are rigorous monotones that evaluate non-classical features such as entanglement, coherence, and magic across quantum systems.
• They employ methods like regularized relative entropy and robustness measures, leveraging convex geometry to link operational advantages with conversion limits.
• These quantifiers underpin practical applications in quantum information processing by dictating allowed state conversions and optimizing discrimination and metrological tasks.

Quantum information resource quantifiers are the rigorous, task-motivated monotones or measures that underlie quantum resource theories (QRTs) across all levels of quantum systems (states, operations, processes, and measurements). These quantifiers characterize non-classical features such as entanglement, coherence, magic, asymmetry, and operationally relevant advantages, providing both geometric and operational metrics for the amount of resource present in a quantum object and dictating the allowed conversion rates via free operations. Their design and properties reveal the fundamental structure of quantum information and the ultimate limits on information processing capabilities.

1. Foundations of Quantum Resource Quantification

In the general structure of a QRT, one first specifies a set of free states $\mathcal{F}\subset\mathcal{D}(\mathcal{H})$. For a given resource %%%%1%%%%, a quantifier $M(\rho)$ should be:

• Faithful: $M(\rho)=0$ if and only if $\rho\in\mathcal{F}$.
• Monotonic: $M(\Phi(\rho))\leq M(\rho)$ for all free operations $\Phi$.
• Additive (asymptotically): $M(\rho^{\otimes n})\sim nM(\rho)$ for large $n$.

For reversible theories with convex, closed $\mathcal{F}$ and maximal free operations (those that do not generate resource, even asymptotically), the central result is that the regularized relative entropy of resource is the unique asymptotic quantifier of resource. All other monotones satisfying the fundamental properties collapse to this measure in the many-copy limit (Brandão et al., 2015).

Regularized Relative Entropy of Resource

For a state $\rho$,

$$E_R(\rho) := \inf_{\sigma\in\mathcal{F}} S(\rho\|\sigma)$$

where $$S(\rho\|\sigma) = \operatorname{Tr}[\rho(\log\rho - \log\sigma)]$$. In the asymptotic regime,

$$E_R^{\infty}(\rho) := \lim_{n\to\infty} \frac{1}{n} E_R(\rho^{\otimes n})$$

is the unique, additive "resource content" per copy in reversible QRTs (Brandão et al., 2015).

Robustness and Logarithmic Robustness

The global robustness is

$$R(\rho) := \min\{s\geq 0 : \exists\tau\in\mathcal{D}(\mathcal{H}),\, \frac{\rho+s\tau}{1+s}\in\mathcal{F}\}$$

with logarithmic robustness

$LR(\rho) := \log[1 + R(\rho)]$

The smoothed log-robustness, accounting for $\varepsilon$-perturbations, is

$$R_L^\varepsilon(\rho) := \min_{\tilde{\rho}\,:\, \|\tilde{\rho}-\rho\|_1 \leq \varepsilon} \log[1+R(\tilde{\rho})]$$

In the asymptotic limit, the regularized relative entropy equals the (smoothed) logarithmic robustness rate: $$E_R^{\infty}(\rho) = \lim_{\varepsilon\to 0} \lim_{n\to\infty} \frac{1}{n} LR^\varepsilon(\rho^{\otimes n})$$ (Brandão et al., 2015).

2. Robustness Measures: Geometry, Duality, and Operational Meaning

Robustness quantifiers unify geometric and operational approaches:

• They are defined as conic optimization tasks (SDPs or infinite-dimensional analogs) (Uola et al., 2018, Lami et al., 2020).
• Dual formulations yield resource witnesses, with the optimal value in the dual corresponding to the maximal advantage in discrimination tasks.
• They possess convexity, strong monotonicity, and (where applicable) lower semicontinuity (Haapasalo et al., 2020, Lami et al., 2020, Regula, 2017).

For a convex, closed free set, the general (or generalized) robustness quantifies the minimal mixing required with any (possibly resourceful) noise to erase the resource. In infinite dimensions, these definitions extend using Banach space duality, with special consideration for topology to maintain faithfulness and operational meaning (Haapasalo et al., 2020, Lami et al., 2020). The robustness always quantifies the maximal possible relative advantage in some discrimination or "quantum game" task (Uola et al., 2018, Haapasalo et al., 2020, Lami et al., 2020).

Examples:

Resource	Free Set $\mathcal{F}$	Robustness Formula	Dual/Operational Task
Entanglement	separable states	$R_{\rm sep}(\rho)$	Channel discrimination advantage (Uola et al., 2018)
Coherence	diagonals in a basis	$R_{\rm coh}(\rho)$	Phase discrimination (Uola et al., 2018)
Nonclassicality	coherent states in optics	$R_{\rm nc}(\ket{n}) = e^n n!/n^n$	Optimal witness, game advantage (Lami et al., 2020)
Magic	stabilizer states	$R_{\rm magic}(\rho)$	Indicator of Clifford non-simulability (Kožić et al., 10 Feb 2025)

The logarithmic and smoothed versions directly connect robustness with the asymptotic resource rate $E_R^\infty$ (Brandão et al., 2015).

3. Convex Geometry and Hierarchies of Resource Quantifiers

A unified convex-geometric framework accommodates norm-based, convex-roof, gauge-functional, and witness-based resource quantifiers:

• For a resource theory with pure free set $V$, quantifiers include:
  • Vector gauge (quantifies minimal decomposition in $V$).
  • Matrix gauges (base-norm, nuclear gauge, convex-roof extensions).
  • Robustness measures as Minkowski functionals.
• Robustness measures and certain atomic gauges are equal on pure states, and all quantifiers admit efficient dual (witness) forms (Regula, 2017).

This framework enables the construction of hierarchies—e.g., $k$-level coherence, $k$-Schmidt entanglement, $k$-partite entanglement, and stabilizer (magic) resources—by varying the choice of $V$ (Regula, 2017).

Pure-state simplification: For $|\psi\rangle$, all convex-geometric quantifiers collapse to the squared vector gauge, e.g.,

$$R(|\psi\rangle\langle\psi|) = \mu_V(|\psi\rangle)^2 - 1$$

for an appropriate choice of $V$.

4. Extensions: Channels, Operations, Measurements, and Processes

Quantum Channels

Channel resource quantifiers generalize state divergence measures:

• Numerous asymptotic and one-shot relative entropy quantifiers: $D_A$, $D_{\Phi^+}$, $D_{AB}$, with $S_{AB}$ being fully monotonic under superchannels and additive (Yuan, 2018).
• Channel robustness and resource measures can be constructed by minimizing over free channel sets.
• Channel-relative entropy of coherence and entanglement admits operational and convex programming formulations.

Measurements and Assemblages

Resource monotones for sets of POVMs use geometric (distance-based) measures:

• The diamond norm distance between measure-and-prepare channels encapsulates operational discrimination power (Tendick et al., 2022).
• A monotone $R_\diamond$ is the minimal diamond norm distance to the free set (unbiased, incoherent, or compatible measurements).
• Hierarchies are established: informativeness $\geq$ coherence $\geq$ incompatibility $\geq$ steering $\geq$ Bell nonlocality (Tendick et al., 2022).

Dynamical Resources, Processes, and Combs

For multitime processes (quantum combs), resource monotones include temporal mutual information quantifiers, corrected for noise reducibility:

• Quantifiers $$\overline{I}_{\hat m}, \overline{M}_{\hat m}, \overline{N}_{\hat m}$$ express the maximal extractable mutual information (total, Markovian, non-Markovian) via independent quantum instrument (IQI) controls (Berk et al., 16 Oct 2025).
• These quantities are monotones under free operations on the process and satisfy intuitive composition rules, with operational significance for the practical utility of temporal correlations.

5. Operational and Information-Theoretic Quantifiers

Operationally meaningful quantifiers occupy a central role:

• Asymptotic conversion rate: For states $\rho$, $\sigma$, under maximal free operations, the optimal rate is

$$R(\rho\to\sigma) = \frac{E_R^\infty(\rho)}{E_R^\infty(\sigma)}$$

(Brandão et al., 2015).

• Quantum Fisher information (QFI): Both classical and quantum Fisher informations vanish on $\mathcal{F}$ and become strictly positive on resourceful states, providing universally faithful resource witnesses and connecting quantum resources to metrological advantage. The maximum metrological gain is tightly bounded in terms of the robustness squared (Tan et al., 2021).
• Intrinsic steerability via conditional mutual information (CMI): A CMI-based monotone $S(\overline{A};B)$ quantifies deviation from local-hidden-state models for steering assemblages, satisfying faithfulness, convexity, and monotonicity under 1W-LOCC, and capturing operational costs (e.g., in measurement simulation) (Kaur et al., 2016).

6. Resource Quantifiers in Many-Body and Computational Settings

Advanced numerical methods, such as tensor cross interpolation, enable the scalable computation of non-linear resource quantifiers (e.g., stabilizer Rényi entropies for magic, relative entropy of coherence) in complex many-body systems:

• These algorithms approximate high-dimensional resource tensors with controllable sampling complexity, allowing quantification of resource content in ground states of critical and ordered models, thus connecting many-body physics to QRTs (Kožić et al., 10 Feb 2025).
• Similar frameworks accommodate arbitrary resource monotones admitting a tensor expansion, including entanglement, discord, and generalized robustness.

7. Consistency, Uniqueness, and Trade-off Criteria

For general theories (including those lacking a unique maximally resourceful state), a consistent resource measure $R$ must satisfy

$R_D(\rho) \leq R(\rho) \leq R_C(\rho)$

where $R_D$ is the distillable resource and $R_C$ is the resource cost, both defined with respect to rates of asymptotic conversion to and from maximally resourceful states. This uniqueness-type bound is foundational even in infinite-dimensional or nonconvex settings, and accommodates the possibility of catalytically replicable (infinitely amplifiable) states (Kuroiwa et al., 2020).

Fundamental trade-offs—expressed as exclusion principles or resource complementarity—emerge in operational contexts:

• In finite dimensions, a central constraint for tripartite systems is:

$q_1^2 + q_2^2 + q_3^2 \leq 1$

where $q_1$ is teleportation advantage, $q_2$ is optimal cloning fidelity, and $q_3$ is normalized quantum Fisher information (coherence-based metrology). The sum $\mathcal{I}=q_1^2+q_2^2+q_3^2$ defines a dynamical invariant (for symmetry-preserving evolutions) and contracts irreversibly under decoherence (Edmondson, 13 Jun 2025).

• M. B. Plenio and S. S. Virmani, "General structure of quantum resource theories," (Brandão et al., 2015).
• T. Haapasalo et al., "An operational characterization of infinite-dimensional quantum resources," (Haapasalo et al., 2020).
• P. Regina et al., "Framework for resource quantification in infinite-dimensional general probabilistic theories," (Lami et al., 2020).
• B. Regula, T. Theurer, and G. Adesso, "Convex geometry of quantum resource quantification," (Regula, 2017).
• M. Takagi and G. Gour, "General Quantum Resource Theories: Distillation, Formation and Consistent Resource Measures," (Kuroiwa et al., 2020).
• A. Edmondson, "Quantum Resource Complementarity in Finite-Dimensional Systems," (Edmondson, 13 Jun 2025).
• L. Tindall et al., "Computing Quantum Resources using Tensor Cross Interpolation," (Kožić et al., 10 Feb 2025).
• E. Chitambar and G. Gour, "Quantifying quantum resources with conic programming," (Uola et al., 2018).
• X. Yuan, "Relative entropies of quantum channels with applications in resource theory," (Yuan, 2018).
• C. El Bakraoui et al., "Enhancing the estimation precision of an unknown phase shift in multipartite Glauber coherent states via skew information correlations and local quantum Fisher information," (Bakraoui et al., 2021).
• Y. Liu et al., "Conditional Mutual Information and Quantum Steering," (Kaur et al., 2016).
• P. W. Shor, "Distance-based resource quantification for sets of quantum measurements," (Tendick et al., 2022).