pith. sign in

arxiv: 2606.21778 · v1 · pith:66YIGUCFnew · submitted 2026-06-19 · 🧮 math.PR

Pontryagin Maximum Principle in Free Probability Theory

Georg Schl\"uchtermann , Michael Wibmer This is my paper

Pith reviewed 2026-06-26 13:05 UTC · model grok-4.3

classification 🧮 math.PR
keywords free probabilityPontryagin maximum principlestochastic controlnon-commutative Ito formularandom matricesfree SDEsoptimal control
0
0 comments X

The pith

The Pontryagin maximum principle extends to optimal control problems in free probability theory.

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

The paper sets up an optimal control problem in free probability consisting of a controlled forward equation together with a free backward stochastic differential equation. It establishes global existence for both equations and defines a Hamiltonian that incorporates the non-commutative Ito formula. The main result is a proof that this Hamiltonian yields the same maximum-principle characterization of optimality that holds in the classical commutative setting. The framework is illustrated by examples in which the principle produces explicit solutions.

Core claim

We formulate an optimal control problem in the setting of free probability, consisting of the controlled forward equation, a free backward stochastic differential equation. For both, we give global existence theorems. Due to the non-commutative Ito-formula, the definition of the Hamiltonian differs to the commutative case. Our strategy is to stay as close as possible to the commutative case. Finally we formulate and proof the maximum principle in the context of free probability.

What carries the argument

The Hamiltonian constructed from the non-commutative Ito formula applied to the controlled free SDE and the free BSDE.

If this is right

  • Global existence holds for the controlled forward equation and the free backward SDE.
  • The maximum principle supplies necessary conditions for optimality.
  • Explicit optimal controls can be obtained from the principle in concrete examples.

Where Pith is reading between the lines

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

  • The same construction may apply directly to optimization tasks formulated on large random matrices.
  • It could supply necessary conditions for problems in free stochastic calculus beyond the examples treated.
  • The approach opens the possibility of deriving optimality conditions for other non-commutative processes that admit an Ito formula.

Load-bearing premise

The non-commutative Ito formula permits a Hamiltonian definition that supports the same optimality characterization as in the commutative case, allowing the proof strategy to carry over after suitable adjustments.

What would settle it

A specific controlled free SDE together with a candidate control that satisfies the maximum-principle condition yet fails to be optimal.

read the original abstract

Motivated by the classical stochastic maximum principle, random matrices and free stochastic differential equations we, develop an analog maximum principle for control problems driven by non-commutative random variables, e.g. random matrices. We formulate an optimal control problem in the setting of free probability, consisting of the controlled forward equation, a free backward stochastic differential equation. For both, we give global existence theorems. Due to the non-commutative It\^{o}-formula, the definition of the Hamiltonian differs to the commutative case. Our strategy is to stay as close as possible to the commutative case. Finally we formulate and proof the maximum principle in the context of free probability. Several examples show the application of the maximum principle, where explicit solutions can be found.

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.

Referee Report

0 major / 3 minor

Summary. The paper formulates an optimal control problem in free probability via a controlled forward free SDE and an associated free backward SDE, establishes global existence theorems for both equations, defines a modified Hamiltonian using the non-commutative Itô formula, and proves a Pontryagin-type maximum principle by adapting the classical commutative strategy while remaining close to it. Several examples are given in which explicit solutions are obtained via the principle.

Significance. If the existence theorems and the maximum-principle proof are rigorous, the work supplies a non-commutative counterpart to the stochastic maximum principle that is directly applicable to random-matrix and free-probability models. The explicit global-existence results and the concrete examples constitute verifiable contributions even if the optimality characterization requires only modest adjustments from the commutative case.

minor comments (3)
  1. [§3 or §4] The abstract states that global existence theorems are proved for the forward and backward equations, but the manuscript should include a brief comparison (e.g., in §3 or §4) with the corresponding commutative results to clarify which estimates carry over unchanged and which require new arguments due to non-commutativity.
  2. [Hamiltonian definition section] Notation for the non-commutative Itô formula and the resulting Hamiltonian should be introduced with an explicit side-by-side display against the classical commutative Hamiltonian (perhaps as a displayed equation in the section defining the Hamiltonian) to make the modification transparent.
  3. [Examples section] In the examples, the optimality condition obtained from the maximum principle should be cross-checked against a direct variational computation or an explicit solution of the controlled SDE when possible, to confirm that the free-probability version recovers the expected optimizer.

Simulated Author's Rebuttal

0 responses · 0 unresolved

We thank the referee for the positive summary of our manuscript, the assessment of its significance, and the recommendation for minor revision. The referee's description accurately captures the formulation of the controlled forward free SDE, the free backward SDE, the global existence results, the modified Hamiltonian, and the proof strategy for the Pontryagin-type maximum principle.

Circularity Check

0 steps flagged

No significant circularity detected

full rationale

The paper formulates an optimal control problem via controlled forward free SDE and free backward SDE, proves global existence for both, defines a modified Hamiltonian using the non-commutative Itô formula, and establishes the maximum principle by adapting the classical commutative strategy while staying close to it. No load-bearing step reduces by construction to a fitted parameter, self-definition, or self-citation chain; the derivation draws on external classical results and free probability theory without internal circular reductions.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The central claim rests on existence of solutions to the controlled forward equation and free backward SDE plus a suitably defined non-commutative Hamiltonian; these are stated as given without further derivation in the abstract.

axioms (2)
  • domain assumption Global existence theorems hold for the controlled forward equation and the free backward stochastic differential equation.
    Abstract states these theorems are provided as part of the setup.
  • domain assumption The non-commutative Ito formula yields a Hamiltonian whose maximization condition characterizes optimality.
    Abstract notes the Hamiltonian definition differs from the commutative case and is used to formulate the principle.

pith-pipeline@v0.9.1-grok · 5645 in / 1322 out tokens · 31134 ms · 2026-06-26T13:05:23.913048+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

30 extracted references · 18 canonical work pages

  1. [1]

    A. B. Aleksandrov and V. V. Peller. Operator Lipschitz functions.Russian Mathematical Surveys, 71(4):605–702, August 2016.doi:10.1070/rm9729

    work page doi:10.1070/rm9729 2016

  2. [2]

    Anderson, Alice Guionnet, and Ofer Zeitouni.An Introduction to Random Matrices, volume 118 ofCambridge Studies in Advanced Mathematics

    Greg W. Anderson, Alice Guionnet, and Ofer Zeitouni.An Introduction to Random Matrices, volume 118 ofCambridge Studies in Advanced Mathematics. Cambridge University Press, Cambridge, 2010. URL:https://doi.org/10. 1017/CBO9780511801334

    2010

  3. [3]

    Control in hilbert space and first order mean field type problem, 2021

    Alain Bensoussan, Henry Hang Cheung, and Sheung Chi Phillip Yam. Control in hilbert space and first order mean field type problem, 2021. URL:https: //arxiv.org/abs/2108.11237,arXiv:2108.11237. 26

    arXiv 2021

  4. [4]

    Biane, M

    P. Biane, M. Capitaine, and A. Guionnet. Large deviation bounds for matrix Brownian motion.Inventiones Mathematicae, 152(2):433–459, May 2003.doi: 10.1007/s00222-002-0281-4

    work page doi:10.1007/s00222-002-0281-4 2003

  5. [5]

    Philippe Biane and Roland Speicher. Stochastic calculus with respect to free Brownian motion and analysis on Wigner space.Probability Theory and Re- lated Fields, 112(3):373–409, November 1998.doi:10.1007/s004400050194

    work page doi:10.1007/s004400050194 1998

  6. [6]

    Free diffusions, free entropy and free Fisher information.Ann

    Philippe Biane and Roland Speicher. Free diffusions, free entropy and free Fisher information.Ann. Inst. H. Poincar´ e Probab. Statist., 37(5), 2001.doi: 10.1016/S0246-0203(00)01074-8

    work page doi:10.1016/s0246-0203(00)01074-8 2001

  7. [7]

    Boscain, M

    U. Boscain, M. Sigalotti, and D. Sugny. Introduction to the Pontryagin Max- imum Principle for Quantum Optimal Control.PRX Quantum, 2(3), 2021. URL:http://dx.doi.org/10.1103/PRXQuantum.2.030203

    work page doi:10.1103/prxquantum.2.030203 2021

  8. [8]

    Dabrowski, A

    Y. Dabrowski, A. Guionnet, and D. Shlyakhtenko. Free transport for convex potentials, 2016. URL:https://arxiv.org/abs/1701.00132,arXiv:1701. 00132

    Pith/arXiv arXiv 2016

  9. [9]

    A Laplace Principle for Hermitian Brownian Motion and Free Entropy I: the convex functional case, 2017

    Yoann Dabrowski. A Laplace Principle for Hermitian Brownian Motion and Free Entropy I: the convex functional case, 2017. URL:https://arxiv.org/ abs/1604.06420

    Pith/arXiv arXiv 2017

  10. [10]

    A free probabilistic framework for denoising diffusion models: Entropy, transport, and reverse processes, 2025.arXiv:2510.22778

    Swagatam Das. A free probabilistic framework for denoising diffusion models: Entropy, transport, and reverse processes, 2025.arXiv:2510.22778

    arXiv 2025

  11. [11]

    Stochastic optimal control of interacting particle systems in hilbert spaces and applica- tions, 2025

    Filippo de Feo, Fausto Gozzi, Andrzej Swiech, and Lukas Wessels. Stochastic optimal control of interacting particle systems in hilbert spaces and applica- tions, 2025. URL:https://arxiv.org/abs/2511.21646

    arXiv 2025

  12. [12]

    A maximum principle for optimal control of stochastic evolution equations.SIAM Journal on Control and Optimization, 51(6):4343–4362, 2013.doi:10.1137/120882433

    Kai Du and Qingxin Meng. A maximum principle for optimal control of stochastic evolution equations.SIAM Journal on Control and Optimization, 51(6):4343–4362, 2013.doi:10.1137/120882433

    work page doi:10.1137/120882433 2013

  13. [13]

    Wilfrid Gangbo, David Jekel, Kyeongsik Nam, and Aaron Z. Palmer. Viscosity solutions in non-commutative variables, 2025.arXiv:2502.17329

    arXiv 2025

  14. [14]

    Du- ality for optimal couplings in free probability.Communications in Mathemat- ical Physics, 396(3):903–981, Dec 2022.doi:10.1007/s00220-022-04480-0

    Wilfrid Gangbo, David Jekel, Kyeongsik Nam, and Dimitri Shlyakhtenko. Du- ality for optimal couplings in free probability.Communications in Mathemat- ical Physics, 396(3):903–981, Dec 2022.doi:10.1007/s00220-022-04480-0

    work page doi:10.1007/s00220-022-04480-0 2022

  15. [15]

    An elementary approach to free entropy theory for convex poten- tials.Analysis and PDE, 13(8):2289–2374, 2020

    David Jekel. An elementary approach to free entropy theory for convex poten- tials.Analysis and PDE, 13(8):2289–2374, 2020. URL:http://dx.doi.org/ 10.2140/apde.2020.13.2289,doi:10.2140/apde.2020.13.2289

    work page doi:10.2140/apde.2020.13.2289 2020

  16. [16]

    Evolution equations in non-commutative probability.https: //escholarship.org/uc/item/8n39f7mt, 2020

    David Jekel. Evolution equations in non-commutative probability.https: //escholarship.org/uc/item/8n39f7mt, 2020. [Online; accessed 04-March- 2026]

    2020

  17. [17]

    Tracial smooth functions of non-commuting variables and the free Wasserstein manifold.Dissertationes Math., 580, 2022.doi:10.4064/dm843-10-2021

    David Jekel, Wuchen Li, and Dimitri Shlyakhtenko. Tracial smooth functions of non-commuting variables and the free Wasserstein manifold.Dissertationes Math., 580, 2022.doi:10.4064/dm843-10-2021

    work page doi:10.4064/dm843-10-2021 2022

  18. [18]

    On Free Stochastic Differential Equations.Journal of The- oretical Probability, 24(3), 2011.doi:10.1007/s10959-011-0341-z

    Vladislav Kargin. On Free Stochastic Differential Equations.Journal of The- oretical Probability, 24(3), 2011.doi:10.1007/s10959-011-0341-z. 27

    work page doi:10.1007/s10959-011-0341-z 2011

  19. [19]

    Stochastic integration on the Cuntz algebraO ∞.J

    Burkhard K¨ ummerer and Roland Speicher. Stochastic integration on the Cuntz algebraO ∞.J. Funct. Anal., 103(2):372–408, 1992.doi:10.1016/ 0022-1236(92)90126-4

    1992

  20. [20]

    Mingo and Roland Speicher.Free Probability and Random Matrices, volume 35 ofFields Institute Monographs

    James A. Mingo and Roland Speicher.Free Probability and Random Matrices, volume 35 ofFields Institute Monographs. Springer New York, 2017.doi: 10.1007/978-1-4939-6942-5

    work page doi:10.1007/978-1-4939-6942-5 2017

  21. [21]

    Stochastic theta methods for free stochastic differential equations.IMA Journal of Numerical Analysis, 2025.doi:10.1093/imanum/draf044

    Yuanling Niu, Jiaxin Wei, Zhi Yin, and Dan Zeng. Stochastic theta methods for free stochastic differential equations.IMA Journal of Numerical Analysis, 2025.doi:10.1093/imanum/draf044

    work page doi:10.1093/imanum/draf044 2025

  22. [22]

    Springer Berlin Heidelberg, 2009.doi:10.1007/ 978-3-540-89500-8

    Huyˆ en Pham.Continuous-time Stochastic Control and Optimization with Financial Applications. Springer Berlin Heidelberg, 2009.doi:10.1007/ 978-3-540-89500-8

    2009

  23. [23]

    Chapter 34 - Non-Commutative Lp-Spaces

    Gilles Pisier and Quanhua Xu. Chapter 34 - Non-Commutative Lp-Spaces. In W.B. Johnson and J. Lindenstrauss, editors,Handbook of the Geometry of Banach Spaces, volume 2, pages 1459–1517. Elsevier Science B.V., 2003. doi:10.1016/S1874-5849(03)80041-4

    work page doi:10.1016/s1874-5849(03)80041-4 2003

  24. [24]

    Numerical Solution of Free Stochastic Differential Equations.SIAM J

    Georg Schl¨ uchtermann and Michael Wibmer. Numerical Solution of Free Stochastic Differential Equations.SIAM J. Numer. Anal., 61(6), 2023. doi:10.1137/22M1526393

    work page doi:10.1137/22m1526393 2023

  25. [25]

    Springer, Cham, 2019

    Anna Skripka and Anna Tomskova.Multilinear Operator Integrals, volume 2250 ofLecture Notes in Mathematics. Springer, Cham, 2019. Theory and applications.doi:10.1007/978-3-030-32406-3

    work page doi:10.1007/978-3-030-32406-3 2019

  26. [26]

    Free Calculus, 2001

    Roland Speicher. Free Calculus, 2001. URL:https://arxiv.org/abs/math/ 0104004,arXiv:math/0104004

    Pith/arXiv arXiv 2001

  27. [27]

    American Mathematical Society, 2012.doi:10.1090/gsm/132

    Terence Tao.Topics in Random Matrix Theory. American Mathematical Society, 2012.doi:10.1090/gsm/132

    work page doi:10.1090/gsm/132 2012

  28. [28]

    Penghui Wang and Shan Wang. Optimal control of quantum systems in fermion fields: The pontryagin-type maximum principle (i).SIAM Jour- nal on Control and Optimization, 64(3):1357–1382, 2026.arXiv:https: //doi.org/10.1137/24M1644237,doi:10.1137/24M1644237

    work page doi:10.1137/24m1644237 2026

  29. [29]

    On milstein-type methods for free stochastic differential equations, 2026

    Michael Wibmer and Georg Schl¨ uchtermann. On milstein-type methods for free stochastic differential equations, 2026. URL:https://arxiv.org/abs/ 2502.11233,arXiv:2502.11233

    arXiv 2026

  30. [30]

    Springer New York, New York, NY, 1999

    Jiongmin Yong and Xun Yu Zhou.Stochastic Controls. Springer New York, 1999.doi:10.1007/978-1-4612-1466-3. 28

    work page doi:10.1007/978-1-4612-1466-3 1999