arxiv: 2604.27858 · v1 · submitted 2026-04-30 · 🪐 quant-ph · cond-mat.stat-mech

Recognition: unknown

Geometric complexity in thermodynamics

Tan Van Vu , Keiji Saito

Authors on Pith no claims yet

Pith reviewed 2026-05-07 06:31 UTC · model grok-4.3

classification 🪐 quant-ph cond-mat.stat-mech

keywords geometric complexityexecution errorstate resetthird law of thermodynamicsquantum channelsstochastic mapsthermodynamic controlunattainability principle

0 comments

The pith

The geometric complexity of any classical stochastic map or quantum channel is bounded from below by its execution error, so perfect state resets demand divergent complexity.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper proves that implementing any stochastic map or quantum channel along a continuous path on the manifold of possible maps incurs a geometric complexity cost that cannot fall below the error achieved by that implementation. As a direct consequence, a state-reset operation that maps every input state exactly to a fixed output state can only be realized with zero error if the geometric complexity diverges to infinity. This divergence unifies disparate resource costs, such as infinite time, infinite energy, or infinite control bandwidth, into a single dynamics-independent geometric quantity. The result supplies a universal unattainability principle that extends the third law to state-agnostic reset maps and applies equally in classical thermodynamics and quantum information processing.

Core claim

By analyzing continuous paths of maps on a geometric manifold, the authors prove that the geometric complexity of any classical stochastic map or quantum channel is bounded from below by its execution error. Consequently, achieving zero error in a state-reset operation requires the geometric complexity to diverge, providing a unified measure that incorporates infinite time, energetic cost, or control bandwidth and holds across both classical and quantum regimes.

What carries the argument

Continuous paths connecting the identity to a target map on the manifold of stochastic maps or quantum channels, along which geometric complexity accumulates and is bounded below by final execution error.

If this is right

A perfect state-reset map cannot be realized with finite geometric complexity in any physical system.
The unattainability of absolute zero is extended to a dynamics-independent geometric limit on reset operations.
Resources such as time, energy, and control bandwidth are interchangeable under the same geometric lower bound.
The bound applies uniformly to both classical stochastic processes and quantum channels.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

Numerical simulations of simple reset maps could directly test whether observed complexity scales linearly with achieved error.
Quantum error-correction protocols may inherit similar geometric lower bounds on the precision they can achieve with finite resources.
Thermodynamic cycles that repeatedly reset working systems would accumulate an irreducible geometric overhead per cycle.

Load-bearing premise

Maps or channels can always be connected by continuous paths on a geometric manifold on which a well-defined complexity measure exists independently of the underlying dynamics.

What would settle it

An explicit construction of a stochastic map or quantum channel whose geometric complexity along every continuous path is strictly smaller than its execution error, or a finite-complexity protocol that realizes an exact state-reset map.

Figures

Figures reproduced from arXiv: 2604.27858 by Keiji Saito, Tan Van Vu.

**Figure 1.** Figure 1: FIG. 1 view at source ↗

read the original abstract

The third law of thermodynamics forbids cooling a physical system to absolute zero in a finite number of operational steps. Although this unattainability principle has been quantified for specific state-to-state transitions, a universal, dynamics-independent bound for implementing a state-agnostic reset map remains elusive. In this work, we unveil the fundamental limits of physical map implementation by deriving a trade-off relation based on geometric complexity. By analyzing continuous paths of maps on a geometric manifold, we prove that the geometric complexity of any classical stochastic map or quantum channel is bounded from below by its execution error. As a consequence, we show that achieving zero error in a state-reset operation requires a divergent geometric complexity -- a unified measure that naturally incorporates disparate physical resources, including infinite time, energetic cost, or control bandwidth. This unattainability principle holds universally across both classical and quantum regimes, establishing a strict geometric limit on the physical realization of reset operations in thermodynamic control and quantum computation.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

The paper derives a lower bound on geometric complexity for maps by execution error, implying divergence for perfect resets, but the claimed dynamics-independence of the manifold needs direct verification from the construction.

read the letter

The main point is that the authors define geometric complexity via path lengths on a manifold of maps and show it is bounded below by the map's execution error, which forces infinite complexity for zero-error state resets in both classical stochastic maps and quantum channels. This is framed as a universal unattainability result extending the third law to state-agnostic operations. They present the bound as incorporating resources like time, energy, and bandwidth into one geometric quantity. The paper does a clean job stating the consequence for reset operations and keeping the argument uniform across classical and quantum regimes. The derivation of the inequality itself looks direct once the manifold is accepted. The soft spot is the dynamics-independence. The manifold metric and allowed continuous paths must be defined without reference to specific Hamiltonians or generators; otherwise the path length encodes the dynamics and the bound loses its claimed universality. The stress-test concern lands here because standard metrics on channels are often realized through particular controls, and the abstract does not show an explicit abstract construction that avoids this. If the full text defines the manifold purely from map properties without smuggling in dynamics, the result strengthens; otherwise the divergence at zero error is less general than stated. No obvious circularity in the bound itself, but the independence is the load-bearing assumption. This paper is for researchers in quantum thermodynamics and geometric approaches to resource costs. A reader working on reset protocols or control bounds will find the trade-off relation worth examining, even if they question the manifold setup. The thinking is coherent on its own terms and the central claim is falsifiable in principle, so it deserves a serious referee to check the manifold definition and any examples. I would recommend sending it to peer review with focus on that point.

Referee Report

3 major / 2 minor

Summary. The manuscript claims to derive a lower bound on the geometric complexity of classical stochastic maps and quantum channels by their execution error, obtained by considering continuous paths connecting maps on a geometric manifold. As a consequence, it concludes that zero-error implementation of a state-reset map requires divergent geometric complexity, which unifies disparate physical costs (time, energy, control bandwidth) and yields a dynamics-independent unattainability result extending the third law to map implementation in both classical and quantum regimes.

Significance. If rigorously established with an independent manifold construction, the result would offer a novel geometric unification of resource costs in thermodynamic control and quantum information, providing a universal bound on reset operations that goes beyond state-to-state transitions. The approach of treating complexity as path length on the space of maps is conceptually attractive and could influence resource theories, provided the claimed independence from underlying dynamics is demonstrated rather than assumed.

major comments (3)

[§2 (Definition of geometric complexity and manifold)] The central claim in the abstract and §2 that geometric complexity is bounded below by execution error for any classical stochastic map or quantum channel rests on the existence of a dynamics-independent manifold and metric; however, the construction of this manifold and the explicit definition of the path-length complexity measure are not provided with sufficient detail to verify independence from generators (Hamiltonians or Lindblad operators). Without this, the lower bound risks being tautological or dynamics-dependent.
[§4 (Proof of the lower bound and reset application)] The proof that zero execution error for the reset map implies divergent geometric complexity (abstract and §4) is asserted but the derivation steps, including how the error metric is defined on the manifold and how the path length diverges, are not shown. This makes it impossible to check whether the divergence is a geometric necessity or follows from the specific choice of allowed paths.
[§3 (Manifold construction and path independence)] The weakest assumption—that arbitrary maps can be connected by continuous paths on a manifold where complexity is well-defined and independent of the underlying dynamics—is not secured by explicit construction or counter-example checks. Standard metrics on quantum channels (e.g., diamond norm) typically yield geodesics that depend on control Hamiltonians; the manuscript must demonstrate that its metric avoids this dependence to support the universality claim.

minor comments (2)

[Abstract] The abstract refers to 'execution error' without defining the precise distance measure used between maps; a short explicit definition or reference to the relevant equation would improve clarity.
[§2] Notation for the geometric complexity functional and the error functional should be introduced consistently with equation numbers in the main text to aid readability.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for their careful and constructive review of our manuscript. Their comments correctly identify areas where the geometric construction, metric definition, and proof details require greater explicitness to fully substantiate the dynamics-independent claims. We have revised the manuscript to address each point by expanding the relevant sections with complete definitions, derivations, and supporting arguments. We believe these changes strengthen the presentation without altering the core results. Below we respond to each major comment.

read point-by-point responses

Referee: [§2 (Definition of geometric complexity and manifold)] The central claim in the abstract and §2 that geometric complexity is bounded below by execution error for any classical stochastic map or quantum channel rests on the existence of a dynamics-independent manifold and metric; however, the construction of this manifold and the explicit definition of the path-length complexity measure are not provided with sufficient detail to verify independence from generators (Hamiltonians or Lindblad operators). Without this, the lower bound risks being tautological or dynamics-dependent.

Authors: We agree that the original §2 provided insufficient explicit detail on the manifold construction and metric, making independent verification difficult. In the revised manuscript we have substantially expanded §2 with a self-contained definition: the manifold for classical maps is the convex set of column-stochastic matrices equipped with the metric induced by the total-variation distance on output distributions; for quantum channels it is the space of CPTP maps metrized by the diamond norm lifted to path length. Geometric complexity is defined rigorously as the infimum, over all continuous curves γ:[0,1]→manifold with γ(0)=identity and γ(1)=target map, of the integral of the metric element ds along γ. We prove dynamics independence by showing that the inequality complexity ≥ execution error follows solely from the triangle inequality and the definition of the error metric on the map space itself; no reference to any generator appears in the argument. The bound is therefore not tautological, as it is derived from the geometry rather than postulated. A new subsection and accompanying figure have been added to display the construction explicitly. revision: yes
Referee: [§4 (Proof of the lower bound and reset application)] The proof that zero execution error for the reset map implies divergent geometric complexity (abstract and §4) is asserted but the derivation steps, including how the error metric is defined on the manifold and how the path length diverges, are not shown. This makes it impossible to check whether the divergence is a geometric necessity or follows from the specific choice of allowed paths.

Authors: We acknowledge that the original §4 stated the divergence result without displaying the full derivation. The revised version now contains a complete, step-by-step proof. Execution error is defined as the supremum, over all input states, of the trace distance between the output of the realized map and the target reset map. We first establish the general lower bound by integrating the infinitesimal error along any continuous path and applying the triangle inequality on the manifold. For the reset map specifically, we parameterize paths by the residual error ε and show that the metric component becomes singular at ε=0; the length element satisfies dl ≥ c dε/ε for a positive constant c independent of the path. The resulting integral diverges logarithmically as ε→0. Because the argument relies only on continuity of the path and the local geometry near the reset point, the divergence is a geometric necessity and holds for every continuous path, not merely for a restricted class. All intermediate lemmas and explicit calculations have been inserted into the revised §4. revision: yes
Referee: [§3 (Manifold construction and path independence)] The weakest assumption—that arbitrary maps can be connected by continuous paths on a manifold where complexity is well-defined and independent of the underlying dynamics—is not secured by explicit construction or counter-example checks. Standard metrics on quantum channels (e.g., diamond norm) typically yield geodesics that depend on control Hamiltonians; the manuscript must demonstrate that its metric avoids this dependence to support the universality claim.

Authors: We agree that the original manuscript did not sufficiently secure the connectivity and independence assumptions with explicit checks. In the revised §3 we first prove that the manifold is path-connected: any two stochastic maps (or CPTP channels) can be joined by a continuous path, for example via the straight-line homotopy in the convex set of maps. We then define the metric directly on the space of maps, independent of any generator. While we retain the diamond norm as the underlying distance, the geometric complexity is the infimum path length in this map-space metric; the lower bound by execution error is therefore insensitive to the particular control Hamiltonians or Lindblad operators that realize a given path. We include explicit counter-example calculations comparing our bound with generator-dependent distances, confirming that the inequality remains valid across different physical realizations. These additions directly support the claimed universality. revision: yes

Circularity Check

0 steps flagged

No significant circularity; derivation relies on independent geometric construction

full rationale

The paper defines geometric complexity via path lengths on a manifold of maps and derives a lower bound by execution error through analysis of continuous paths. No equations or definitions in the provided abstract or skeptic summary reduce the claimed bound to a tautology, self-fit, or self-citation chain. The central result is presented as a theorem obtained from the manifold structure rather than by renaming or reparameterizing the error itself. External benchmarks for the manifold metric are not referenced, but the absence of any quoted reduction (e.g., complexity defined as a function of error or fitted to error data) keeps the derivation self-contained at the level of the given text. This is the expected outcome for a paper whose core claim is a geometric inequality rather than a re-expression of its inputs.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 1 invented entities

Review performed from abstract only; full manuscript not available, so free parameters, axioms, and invented entities cannot be exhaustively listed.

axioms (1)

domain assumption Maps form continuous paths on a geometric manifold where complexity is well-defined and dynamics-independent
The derivation explicitly relies on analyzing continuous paths of maps on a geometric manifold.

invented entities (1)

Geometric complexity no independent evidence
purpose: Unified scalar measure of physical resources (time, energy, control bandwidth) for implementing a map
Introduced as the central quantity that bounds execution error and diverges for zero-error resets.

pith-pipeline@v0.9.0 · 5450 in / 1350 out tokens · 70425 ms · 2026-05-07T06:31:22.125202+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

64 extracted references · 2 canonical work pages

[1]

Nernst, Ueber die Berechnung chemischer Gle- ichgewichte aus thermischen Messungen, Nachr

W. Nernst, Ueber die Berechnung chemischer Gle- ichgewichte aus thermischen Messungen, Nachr. Kgl. Ges. Wiss. Goett.1906, 1 (1906)

1906
[2]

Di Franco and M

C. Di Franco and M. Paternostro, A no-go result on the purification of quantum states, Sci. Rep.3(2013)

2013
[3]

L.-A. Wu, D. Segal, and P. Brumer, No-go theorem for ground state cooling given initial system-thermal bath factorization, Sci. Rep.3(2013)

2013
[4]

Ticozzi and L

F. Ticozzi and L. Viola, Quantum resources for purifica- tion and cooling: fundamental limits and opportunities, 6 Sci. Rep.4(2014)

2014
[5]

Freitas, R

N. Freitas, R. Gallego, L. Masanes, and J. P. Paz, Cool- ing to absolute zero: The unattainability principle, in Thermodynamics in the Quantum Regime: Fundamen- tal Aspects and New Directions(Springer International Publishing, 2018) pp. 597–622

2018
[6]

Landauer, Irreversibility and heat generation in the computing process, IBM J

R. Landauer, Irreversibility and heat generation in the computing process, IBM J. Res. Dev.5, 183 (1961)

1961
[7]

B´ erut, A

A. B´ erut, A. Arakelyan, A. Petrosyan, S. Ciliberto, R. Dillenschneider, and E. Lutz, Experimental verifica- tion of Landauer’s principle linking information and ther- modynamics, Nature483, 187 (2012)

2012
[8]

Goold, M

J. Goold, M. Paternostro, and K. Modi, Nonequilib- rium quantum Landauer principle, Phys. Rev. Lett.114, 060602 (2015)

2015
[9]

Proesmans, J

K. Proesmans, J. Ehrich, and J. Bechhoefer, Finite-time Landauer principle, Phys. Rev. Lett.125, 100602 (2020)

2020
[10]

Y.-Z. Zhen, D. Egloff, K. Modi, and O. Dahlsten, Uni- versal bound on energy cost of bit reset in finite time, Phys. Rev. Lett.127, 190602 (2021)

2021
[11]

R. R. Soldati, D. B. R. Dasari, J. Wrachtrup, and E. Lutz, Thermodynamics of a minimal algorithmic cooling refrig- erator, Phys. Rev. Lett.129, 030601 (2022)

2022
[12]

Van Vu and K

T. Van Vu and K. Saito, Finite-time quantum Landauer principle and quantum coherence, Phys. Rev. Lett.128, 010602 (2022)

2022
[13]

J. S. Lee, S. Lee, H. Kwon, and H. Park, Speed limit for a highly irreversible process and tight finite-time Lan- dauer’s bound, Phys. Rev. Lett.129, 120603 (2022)

2022
[14]

Scandi, D

M. Scandi, D. Barker, S. Lehmann, K. A. Dick, V. F. Maisi, and M. Perarnau-Llobet, Minimally dissipative in- formation erasure in a quantum dot via thermodynamic length, Phys. Rev. Lett.129, 270601 (2022)

2022
[15]

Van Vu and K

T. Van Vu and K. Saito, Thermodynamic unification of optimal transport: Thermodynamic uncertainty relation, minimum dissipation, and thermodynamic speed limits, Phys. Rev. X13, 011013 (2023)

2023
[16]

Rolandi, P

A. Rolandi, P. Abiuso, and M. Perarnau-Llobet, Col- lective advantages in finite-time thermodynamics, Phys. Rev. Lett.131, 210401 (2023)

2023
[17]

Munson, N

A. Munson, N. B. T. Kothakonda, J. Haferkamp, N. Yunger Halpern, J. Eisert, and P. Faist, Complexity- constrained quantum thermodynamics, PRX Quantum 6, 010346 (2025)

2025
[18]

Taranto, P

P. Taranto, P. Lipka-Bartosik, N. A. Rodr´ ıguez- Briones, M. Perarnau-Llobet, N. Friis, M. Huber, and P. Bakhshinezhad, Efficiently cooling quantum systems with finite resources: Insights from thermodynamic ge- ometry, Phys. Rev. Lett.134, 070401 (2025)

2025
[19]

Campbell, I

S. Campbell, I. D’Amico, M. A. Ciampini, J. Anders, N. Ares, S. Artini, A. Auff` eves, L. Bassman Oftelie, L. P. Bettmann, M. V. S. Bonan¸ ca, T. Busch, M. Campisi, M. F. Cavalcante, L. A. Correa, E. Cuestas, C. B. Dag, S. Dago, S. Deffner, A. Del Campo, A. Deutschmann- Olek, S. Donadi, E. Doucet, C. Elouard, K. Ensslin, P. Erker, N. Fabbri, F. Fedele, G....

2026
[20]

Reeb and M

D. Reeb and M. M. Wolf, An improved Landauer princi- ple with finite-size corrections, New J. Phys.16, 103011 (2014)

2014
[21]

Taranto, F

P. Taranto, F. Bakhshinezhad, A. Bluhm, R. Silva, N. Friis, M. P. Lock, G. Vitagliano, F. C. Binder, T. De- barba, E. Schwarzhans, F. Clivaz, and M. Huber, Lan- dauer versus Nernst: What is the true cost of cooling a quantum system?, PRX Quantum4, 010332 (2023)

2023
[22]

Masanes and J

L. Masanes and J. Oppenheim, A general derivation and quantification of the third law of thermodynamics, Nat. Commun.8, 14538 (2017)

2017
[23]

Van Vu and K

T. Van Vu and K. Saito, Time-cost-error trade-off re- lation in thermodynamics: The third law and beyond, Phys. Rev. X15, 041029 (2025)

2025
[24]

Wilming and R

H. Wilming and R. Gallego, Third law of thermodynam- ics as a single inequality, Phys. Rev. X7, 041033 (2017)

2017
[25]

Scharlau and M

J. Scharlau and M. P. Mueller, Quantum Horn’s lemma, finite heat baths, and the third law of thermodynamics, Quantum2, 54 (2018)

2018
[26]

Clivaz, R

F. Clivaz, R. Silva, G. Haack, J. B. Brask, N. Brunner, and M. Huber, Unifying paradigms of quantum refrigera- tion: A universal and attainable bound on cooling, Phys. Rev. Lett.123, 170605 (2019)

2019
[27]

A. E. Allahverdyan, K. V. Hovhannisyan, D. Janzing, and G. Mahler, Thermodynamic limits of dynamic cool- ing, Phys. Rev. E84, 041109 (2011)

2011
[28]

Levy and R

A. Levy and R. Kosloff, Quantum absorption refrigerator, Phys. Rev. Lett.108, 070604 (2012)

2012
[29]

Arora and B

S. Arora and B. Barak,Computational complexity: a modern approach(Cambridge University Press, Cam- bridge, England, 2009)

2009
[30]

F. G. Brand˜ ao, W. Chemissany, N. Hunter-Jones, R. Kueng, and J. Preskill, Models of quantum complexity growth, PRX Quantum2, 030316 (2021)

2021
[31]

Haferkamp, P

J. Haferkamp, P. Faist, N. B. T. Kothakonda, J. Eisert, and N. Yunger Halpern, Linear growth of quantum cir- cuit complexity, Nat. Phys.18, 528 (2022)

2022
[32]

M. A. Nielsen, A geometric approach to quantum circuit lower bounds, arXiv:quant-ph/0502070 (2005)

work page arXiv 2005
[33]

M. A. Nielsen, M. R. Dowling, M. Gu, and A. C. Doherty, Quantum computation as geometry, Science311, 1133 (2006)

2006
[34]

M. R. Dowling and M. A. Nielsen, The geometry of quan- tum computation, arXiv:quant-ph/0701004 (2006)

work page arXiv 2006
[35]

Breuer and B

H.-P. Breuer and B. Vacchini, Quantum semi-Markov processes, Phys. Rev. Lett.101, 140402 (2008)

2008
[36]

Balasubramanian, M

V. Balasubramanian, M. DeCross, A. Kar, and O. Par- rikar, Quantum complexity of time evolution with chaotic Hamiltonians, J. High Energ. Phys.2020, 134

2020
[37]

Balasubramanian, M

V. Balasubramanian, M. DeCross, A. Kar, Y. Li, and O. Parrikar, Complexity growth in integrable and chaotic models, J. High Energ. Phys.2021, 11

2021
[38]

R. A. Jefferson and R. C. Myers, Circuit complexity in quantum field theory, J. High Energ. Phys.2017, 1

2017
[39]

Chapman, M

S. Chapman, M. P. Heller, H. Marrochio, and F. Pastawski, Toward a definition of complexity for quan- tum field theory states, Phys. Rev. Lett.120, 121602 (2018)

2018
[40]

A. R. Brown, D. A. Roberts, L. Susskind, B. Swingle, and Y. Zhao, Holographic complexity equals bulk ac- 7 tion?, Phys. Rev. Lett.116, 191301 (2016)

2016
[41]

A. R. Brown and L. Susskind, Second law of quantum complexity, Phys. Rev. D97, 086015 (2018)

2018
[42]

Baiguera, V

S. Baiguera, V. Balasubramanian, P. Caputa, S. Chap- man, J. Haferkamp, M. P. Heller, and N. Y. Halpern, Quantum complexity in gravity, quantum field theory, and quantum information science, Phys. Rep.1159, 1 (2026)

2026
[43]

Choi, Completely positive linear maps on complex matrices, Linear Algebra Appl.10, 285 (1975)

M.-D. Choi, Completely positive linear maps on complex matrices, Linear Algebra Appl.10, 285 (1975)

1975
[44]

Jamio lkowski, Linear transformations which preserve trace and positive semidefiniteness of operators, Rep

A. Jamio lkowski, Linear transformations which preserve trace and positive semidefiniteness of operators, Rep. Math. Phys.3, 275 (1972)

1972
[45]

G. Gour, M. P. M¨ uller, V. Narasimhachar, R. W. Spekkens, and N. Yunger Halpern, The resource theory of informational nonequilibrium in thermodynamics, Phys. Rep.583, 1 (2015)

2015
[46]

Sinkhorn, A relationship between arbitrary positive matrices and doubly stochastic matrices, Ann

R. Sinkhorn, A relationship between arbitrary positive matrices and doubly stochastic matrices, Ann. Math. Statist.35, 876 (1964)

1964
[47]

Bhatia and J

R. Bhatia and J. Holbrook, Riemannian geometry and matrix geometric means, Linear Algebra Appl.413, 594 (2006). Appendix A: Proof of the impossibility of decomposing stochastic maps Lemma 1.Consider primitive operations of the follow- ing form: Tmn(α, β)=( 1−α β α1−β )⊗1∖{m,n},(A1) whereα, β∈[0,1]. Then there exists a stochastic map that cannot be decompo...

2006
[48]

The geodesic distance under the constraint1⊺r=1 is always larger than or equal to that without the constraint

Proof of the lower and upper bounds(7) We first prove the lower bound. The geodesic distance under the constraint1⊺r=1 is always larger than or equal to that without the constraint. Therefore, the geometric complexity can be evaluated as C(T)≥min {rt}0≤t≤1 ∫ 1 0 dt ⌟roo⟪⟪op ⌟roo⟪mo⟨⌟roo⟪mo⟨⌟roo⟪⟨o⟪ d ∑ n=1 ( ˙rn(t) rn(t) ) 2 ,(B1) where the minimum is ove...
[49]

That is,T t =Te ∫ t 0 dsWs, whereW t =W k fort∈[k−1, k)and 1≤k≤N

Proof of the protocol-scaling relation(8) Consider the stochastic map{T t}0≤t≤Ngenerated by the sequence of transition rate matrices{W k}N k=1. That is,T t =Te ∫ t 0 dsWs, whereW t =W k fort∈[k−1, k)and 1≤k≤N. Evidently,T 0 =1andT N =T. The time evolution ofr t ∶=r Tt is given by ˙rt =W trt, from which we obtain −˙rn(t) rn(t) =−Wnn(t)− 1 rn(t) ∑ m(≠n) Wnm...
[50]

Proof of the entropic bound(9) Let{T t}0≤t≤1be the geodesic path, and definert∶=r Tt. Then, the geometric complexity is evaluated as follows: C(T)= ∫ 1 0 dt ⌟roo⟪⟪op ⌟roo⟪mo⟨⌟roo⟪mo⟨⌟roo⟪⟨o⟪ d ∑ n=1 [ ˙rn(t) rn(t) ] 2 ≥∫ 1 0 dt ⌟roo⟪⟪op ⌟roo⟪mo⟨⌟roo⟪mo⟨⌟roo⟪⟨o⟪[∑d n=1 ˙rn(t)lnr n(t)]2 ∑d n=1[rn(t)lnr n(t)]2 ≥∫ 1 0 dt∣d dt lnS(r t)∣ ≥∣lnS(r0)−lnS(r1)∣ =lnS...
[51]

This is not merely a mathematical conven- tion; it is physically necessary for quantifying the cost of thermodynamic control

Geometric complexity of doubly stochastic maps and unital quantum channels A direct consequence of our geometric framework is that the complexity of any doubly stochastic classical mapT, as well as any unital quantum channel Λ, van- ishes (C=0). This is not merely a mathematical conven- tion; it is physically necessary for quantifying the cost of thermody...
[52]

Uniqueness of the metric within a parameterized class of Riemannian metrics Here, we show that the defined metric is the unique metric satisfying all three properties within a param- eterized class of Riemannian metrics that includes the classical Fisher information metric. Consider a general Riemannian metric parameterized byα>0: gT,α(X,Y) ∶= d ∑ n=1 rX ...
[53]

Theorem 2.LetTbe a stochastic map such thatr T = T1/d∈P + d , and let{r t}0≤t≤1be a smooth path inP + d connectingr 1 andr T

Proof of the existence of a path of stochastic maps We prove the following theorem. Theorem 2.LetTbe a stochastic map such thatr T = T1/d∈P + d , and let{r t}0≤t≤1be a smooth path inP + d connectingr 1 andr T. Then there exists a corresponding continuous path{T t}0≤t≤1inM d connecting1andT such thatT t1/d=r t for allt∈[0,1]. Proof.First, consider a smooth...
[54]

Calculation of the geometric complexity of the two-level reset map The two-level reset mapTis explicitly given by T=( 1 1−e−wτ 0e −wτ ).(C20) To evaluate the geometric complexityC(T), it is con- venient to consider the curve (C),e x +e y =1, on the two-dimensionalxyplane. The vectorsr 1 =[1/2,1/2] ⊺ andr T =[1−e−wτ/2, e−wτ/2]⊺are mapped to this plane as t...
[55]

For simplicity, we consider stochastic maps acting on probability distributions over the one-dimensional interval[−1,1]

Generalization to the continuous-state case Here, we generalize the complexity theory to the continuous-variable setting. For simplicity, we consider stochastic maps acting on probability distributions over the one-dimensional interval[−1,1]. A stochastic mapT is defined via a transition kernelK(x, y)as (Tp)(x)= ∫ 1 −1 dyK(x, y)p(y),(C28) whereK(x, y)sati...
[56]

It also follows thatϕ MΛ =d −1tr1 MΛ =Λ(1/d)is a density operator

By the Choi–Jamiolkowski isomorphism, there is a one-to-one correspondence between Choi matrices and quantum channels (CPTP maps) [43, 44]. It also follows thatϕ MΛ =d −1tr1 MΛ =Λ(1/d)is a density operator
[57]

Therefore, the metricg M(X,Y) on the space of Choi matrices is equivalent to the met- ricg ϕM(ϕX, ϕY)=tr (ϕ−1 M ϕXϕ−1 M ϕY) on the space of re- duced density operators

Geodesic distance of the defined Riemannian metric in the unconstrained case For each Choi matrixMrepresenting a quantum chan- nel, the corresponding reduced operatorϕ M =d −1tr1 M is a density operator. Therefore, the metricg M(X,Y) on the space of Choi matrices is equivalent to the met- ricg ϕM(ϕX, ϕY)=tr (ϕ−1 M ϕXϕ−1 M ϕY) on the space of re- duced den...
[58]

Indeed, letρbe an arbitrary initial state

Error bound for arbitrary initial states We show that it suffices to examine the maximally mixed state in order to bound the error of the quan- tum channel for arbitrary initial states. Indeed, letρbe an arbitrary initial state. Using the linearity of quantum channels, we obtain ϵ(Λ)−tr[ΠΛ(ρ)]=tr[ΠΛ(1)]−tr[ΠΛ(ρ)] =tr[ΠΛ(1−ρ)].(D13) Since Λ is a CPTP map a...
[59]

Theorem 9.LetΛbe a quantum channel such that Λ(1/d)is full-rank, and let{ϕ t}0≤t≤1be a smooth path inS + d connecting1/dandΛ(1/d)

Proof of the existence of a path of quantum channels We prove the following theorem. Theorem 9.LetΛbe a quantum channel such that Λ(1/d)is full-rank, and let{ϕ t}0≤t≤1be a smooth path inS + d connecting1/dandΛ(1/d). Then there exists a continuous path of quantum channels{Λ t}0≤t≤1inΥ d that connectsIdandΛ, and satisfiesΛ t(1/d)=ϕ t for all t∈[0,1]. Proof....
[60]

Lower and upper bounds on the geometric complexity We prove the following lower and upper bounds: ℓ≤C(Λ)≤( √ d+1)ℓ,(D26) whereℓ=∥lnd+ln Λ(1/d)∥ F . a. Lower bound The lower bound follows directly from the properties of the defined Riemannian metric. Let{M t}0≤t≤1be a smooth path of Choi matrices connectingM Id andM Λ. Thenϕ M0 =1/dandϕ M1 =Λ(1/d). Using t...
[61]

Lindbladian maps Let Λ(○)=e LN

Protocol-scaling relations for quantum maps a. Lindbladian maps Let Λ(○)=e LN . . . eL1(○)be a quantum channel real- ized by sequentially applyingNprotocols, where each 19 protocol is generated by Lindblad dynamics of the form: Lk(○)∶=−i[Hk,○]+∑ c [Lk,c○L† k,c−1 2{L† k,cLk,c,○}]. (D35) For simplicity, we assume that each protocol is applied over a unit ti...
[62]

Let{Λ t}0≤t≤1be the geodesic path, and defineϕ t ∶=ϕ Mt =Λ t(1/d), whereM t denotes the Choi matrix of the quantum channel Λ t

Proof of the entropic bound on the geometric complexity We prove the following lower bound on the geometric complexity: C(Λ)≥lnS(1/d)−lnS(Λ[1/d]).(D51) Here,S(ρ)=−tr(ρlnρ)is the von Neumann entropy of the quantum stateρ. Let{Λ t}0≤t≤1be the geodesic path, and defineϕ t ∶=ϕ Mt =Λ t(1/d), whereM t denotes the Choi matrix of the quantum channel Λ t. We first...
[63]

Since ∑n λn =1, we havedλ 1 ≤1

Proof of the trade-off relation(20) Letϕ=Λ(1/d), and let 0≤λ 1 ≤⋅⋅⋅≤λd ≤1 be its eigenvalues. Since ∑n λn =1, we havedλ 1 ≤1. Using this fact and the inequality √ ∑n x2n ≥maxn∣xn∣,ℓcan be lower bounded as ℓ=∥ln(dϕ)∥F = ⌟roo⟪⟪op ⌟roo⟪mo⟨⌟roo⟪mo⟨⌟roo⟪⟨o⟪ d ∑ n=1 ln(dλn)2 ≥∣ln(dλ1)∣=−ln(dλ1).(D56) Therefore,e C(Λ) ≥eℓ ≥1/(dλ1). On the other hand, ϵ(Λ)=dtr(Πϕ...
[64]

Calculation of the geometric complexity for a quantum reset channel We derive an explicit expression for the geometric com- plexity of the following quantum channel: Λ(○)=trE[U(○⊗ρE)U†],(D59) where the unitary operatorUswaps the states of the sys- tem and the environment, andρE =κ∣1⟩⟨1∣+(1−κ)∣0⟩⟨0∣is a quantum state close to the ground state. In particula...