Pith. sign in

REVIEW 2 major objections 4 minor 2 cited by

Controlled Occupied Processes and Viscosity Solutions

T0 review · 2 major / 4 minor · reviewed 2026-08-12 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read The value function of an occupied-process control problem is the unique viscosity solution of its associated PDE, by a new comparison principle on the space of positive measures.

desk verdict The comparison machinery is a real step forward, but the stated uniqueness theorem is not supported by the bounded comparison result — without a growth class, the occupied heat equation has nontrivial Tychonoff solutions. read the letter →

arxiv 2411.12080 v1 pith:N56NEY5E submitted 2024-11-18 math.OC math.AP

H. Mete Soner , Valentin Tissot-Daguette , Jianfeng Zhang This is my paper · ORCID
classification math.OCmath.AP MSC 49L1235K5535R1560J5593E20
keywords StochasticoptimalcontroloccupationflowoccupiedPDEsviscositysolutionscomparisonprincipleCrandall-Ishii-Lionslemmainfinite-dimensionalPDEmeasure-valuedstate
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper proves a comparison principle for viscosity solutions of the occupied PDE, the infinite-dimensional dynamic programming equation of stochastic control problems whose state includes the occupation measure $O_t=\int_0^t\delta_{X_s}\,d\Lambda_s$. The main theorem shows that any upper semicontinuous bounded-above subsolution and lower semicontinuous bounded-below supersolution, ordered on the boundary, are ordered on the whole domain. A direct corollary is that the value function of the control problem is the unique viscosity solution. This matters because it gives a nearly classical viscosity framework for a large class of path-dependent control problems, including pricing PDEs for timer options, Asian payoffs, and uncertain-volatility models, without treating full path-dependence.

What carries the argument

The argument is carried by three interlocking objects: the occupied process $(O_t,X_t)$ with $O_t$ a measure-valued occupation flow; the occupation derivative $\partial_o\varphi(o,x)=\lim_{h\downarrow0}(\varphi(o+h\delta_x,x)-\varphi(o,x))/h$, which replaces the time derivative in parabolic PDEs; and a cylindrical norm $\rho(o,x)=(\sum_k|o(f_k)|^2+|x|^2)^{1/2}$ built from a separating family $(f_k)$ satisfying $\sum_k\|f_k\|^2_{C^1}\le1$. With this norm, the paper projects the infinite-dimensional problem onto finite-dimensional subspaces $\pi_K(o)=(o(f_1),\dots,o(f_K))$, proves a Crandall-Ishii-Lions lemma (Lemma 5.1) producing second-order jet inequalities in the limit, and uses the Hamiltonian's Crandall-Ishii-Lions property (Lemma 6.2) to run a doubling-of-variables comparison with coercive approximations $w_\gamma=w+\gamma\vartheta$, $u_\gamma=u-\gamma\vartheta$. The vanishing-penalty limit $\varepsilon\downarrow0$ then yields the contradiction proving comparison.

What would settle it

Take $d=1$ and a separating family $(f_k)$ satisfying (2.4) whose span is not dense in $C_b(\mathbb{R})$, and check whether $\rho(o_n-o)\to0$ with $o_n,o\in M_T$ forces $o_n\to o$ weakly. If a counterexample exists, then the claim that $\rho$ metrizes the weak topology on $M_T$ fails, and the comparison proof's reliance on $\rho$ as the metric of the state space needs a density assumption on $(f_k)$ to survive.

Watch

Extended reading notes

Core claim

Under Assumption 3.4 (growth, Lipschitz, and a nondegeneracy condition on the clock rate $\lambda$), the paper establishes Theorem 3.9: if $u$ is a viscosity subsolution of $H(o,x,\partial_o u,\nabla u,\nabla^2 u)=0$ that is upper semicontinuous and bounded above, $w$ is a viscosity supersolution that is lower semicontinuous and bounded below, and $u\le w$ on the boundary $\partial D_T$, then $u\le w$ on $D_T$. The proof uses coercive approximations to overcome the non-compactness of the measure space, an infinite-dimensional Crandall-Ishii-Lions lemma obtained by finite-dimensional projection, and a Crandall-Ishii-Lions property of the Hamiltonian. Theorem 3.10 then identifies the value function $v(o,x)=\inf_\alpha J(o,x,\alpha)$ as the unique viscosity solution, extending classical stochastic-control uniqueness to state spaces of the form $M_q\times\mathbb{R}^d$.

Load-bearing premise

The proof assumes that the cylindrical distance between the two penalized maximizers collapses as the penalty parameter $\varepsilon$ goes to zero—the vanishing-penalty identity $(1/\varepsilon)\rho^2(o_\varepsilon-\bar o_\varepsilon,x_\varepsilon-\bar x_\varepsilon)\to0$—and that the cylindrical norm $\rho$ metrizes the weak topology on the non-compact set $M_T$; both are asserted rather than fully established, and neither is automatic because $M_q$ is incomplete and $D_T$ is not compact.

Editorial extensions

If this is right

  • The value function $v$ is the unique viscosity solution of the occupied PDE, so the dynamic programming equation is a complete characterization of the optimal-control problem, not just a necessary condition.
  • Comparison holds for fully nonlinear Hamiltonians on the space of positive measures, covering the occupied heat equation and the pricing PDEs of timer options, Asian-style payoffs, and uncertain-volatility exotic options.
  • Path dependence enters only through a first-order occupation derivative, so the proof uses classical second-order finite-dimensional tools; it does not require the heavy machinery developed for fully path-dependent PDEs.
  • The infinite-dimensional Crandall-Ishii-Lions lemma and the coercive-approximation technique provide reusable tools for viscosity theory on other non-compact measure spaces.
  • Because the state space is locally compact with coercive sub-levels, the comparison proof avoids Ekeland-type variational principles.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • A testable extension: the same projection scheme should yield comparison for other HJB equations on measure spaces, such as controlled McKean-Vlasov dynamics, provided one verifies the cylindrical-norm metrization and the vanishing-penalty step in that setting; the paper does not prove those transfer.
  • An implicit numerical consequence: with uniqueness in hand, monotone finite-difference or semi-Lagrangian schemes for the occupied PDE, discretizing $o$ by finitely many test functions $f_k$, should converge to the value function, making path-dependent controls computable through measures rather than full paths.
  • The paper's Remark 7.1 conjecture, that the occupied heat equation is classically solvable for every weakly continuous terminal datum, suggests that occupation erasing chronology may yield regularity that path-dependent PDEs lack; if confirmed, it would strengthen the case for the occupied-state formulation.
  • A boundary probe: because the value function is only $1/2$-Hölder in $x$ (Remark 3.6), one can search for an example dropping the nondegeneracy $\lambda\ge1/c_*$ where the comparison theorem fails, delimiting how sharp Assumption 3.4 is.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

2 major / 4 minor

Summary. The paper develops a viscosity theory for parabolic PDEs associated with controlled occupied processes, whose state variable is a measure-valued occupation flow together with a finite-dimensional spatial component. The main technical result is a comparison principle (Theorem 3.9) for bounded viscosity sub- and supersolutions, obtained through an infinite-dimensional Crandall-Ishii-Lions lemma, finite-dimensional projections, and coercive approximations. The authors then claim in Theorem 3.10 that the control problem's value function is the unique viscosity solution of the dynamic programming equation, and they illustrate the framework with the occupied heat equation, timer options, and an uncertain-volatility model. The paper contains complete proofs of an Itô formula and of local 1/2-Hölder regularity of the value function. The central difficulty is that the uniqueness statement is made without any growth class, while the comparison theorem only applies to bounded functions; under the stated linear-growth assumptions the value function can be unbounded, and the uniqueness claim is false as stated.

Significance. If restricted to an appropriate class of solutions, the paper's comparison machinery is a meaningful extension of second-order viscosity theory to a measure-valued state variable. The finite-dimensional projection method, the coercive approximation via the gauge o(q)+q(x), and the complete appendices supplying the Itô formula and the local Hölder regularity are clear strengths. The examples are interesting and well chosen. However, the flagship uniqueness theorem substantially overclaims: no growth condition appears in Definition 3.7 or Theorem 3.10, and the proof of Theorem 3.10 invokes a comparison principle that is proved only for bounded functions. Since the stated assumptions allow unbounded value functions and in fact admit extra unbounded viscosity solutions even for the occupied heat equation, the uniqueness result cannot stand in its present form. The comparison proof also contains an unjustified infinity-limit step. These issues are load-bearing and must be resolved before the manuscript can be considered for publication.

major comments (2)
  1. [Section 3.3, Theorem 3.10; Section 7.1] Theorem 3.10 asserts that the value function v is the unique viscosity solution of (3.12a)-(3.12b), but no growth class is imposed. Theorem 3.9 compares only an upper-bounded subsolution and a lower-bounded supersolution, and the proof uses boundedness essentially in Step 1 and Step 4. Under Assumption 3.4 the payoff data have only linear growth, so v is typically unbounded; hence Theorem 3.9 cannot be applied to two arbitrary viscosity solutions. This is not a mere proof gap: for the occupied heat equation (7.2a)-(7.2b) with g=0, any nonzero Tychonoff solution U(t,x) of the classical heat equation with U(0,·)=0 produces u(o,x)=U(T-|o|,x), a smooth nonzero viscosity solution of -∂_o u - (1/2)Δu=0 with u=0 on ∂D_T. Thus Theorem 3.10 is false as stated. The theorem must be reformulated, either by adding boundedness of ℓ and g (and then proving v is bounded) or by proving and stating a comparison theorem in a specified growth class that excludes Tychonoff-type solutions.
  2. [Theorem 3.9, Step 4, equations (6.5)-(6.6)] The final limiting argument is not justified as written. Equation (6.6) defines c(γ1)=sup_{ε,γ2∈(0,β0]} Q(x_ε^γ,x̄_ε^γ) and claims this is finite. But the coercivity bound derived from u^γ-w^γ yields only Q≤C/γ2, so the supremum over γ2∈(0,β0] can be infinite; for fixed γ1, Q can grow like 1/γ2 as γ2↓0. Consequently the asserted limits lim_{γ1↓0}(γ1 Q²)=0 and, more importantly, lim_{γ2↓0}lim_{γ1↓0}lim_{ε↓0} γ2 Q=0 are not established. The citation to [13, Lemma 3.1] covers the finite-dimensional vanishing-penalty statement, not these limits on the unbounded domain D_T with a measure-valued variable. Since the contradiction 0≤-c0β0 at the end of Step 4 depends on these terms vanishing, the comparison proof has a gap that must be closed, either by a complete limiting argument or by adding a hypothesis that controls the behavior of u and w at infinity.
minor comments (4)
  1. [Section 2.2] The assertion that the cylindrical norm ρ metrizes the weak topology on M_T is false. For example, with f_k∈C_0 chosen as a separating family, the sequence of Dirac measures δ_n satisfies ρ(δ_n,0)→0 but δ_n does not converge weakly to 0 because it is not tight. The paper does not appear to rely on this metrizability for the main proof, since the coercivity of ϑ and the upper semicontinuity of -ρ² suffice for the maximization arguments, but the statement should be corrected or removed.
  2. [Lemma 5.1, Step 4] The displayed formulas for θ_K and θ̄_K write (o_*-ō_*)(f_k), but the definitions in Step 2 and the convergence argument in Step 3 give (o_K-ō_K)(f_k). The limit to Θ(x_*) is correct after a dominated-convergence justification, but the displayed formula should use the K-dependent measures.
  3. [Lemma 6.2, estimate for I1+I2] The expression "|θ| + θ|" should read "|θ| + |θ̄|", and the intermediate bound for the term involving |θ| needs to be displayed consistently with the final ζ_ε definition.
  4. [Theorem 3.9, Step 3] The claim that, for sufficiently small γ and ε, all maximizers of Φ_ε^γ lie in (˚D_T)² is stated without proof. A brief argument showing that a boundary maximizer would make Φ_ε^γ ≤ o(1) as ε↓0, contradicting the positive supremum from (6.2), would make the proof self-contained.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the comparison principle is derived from stated assumptions and the classical finite-dimensional Crandall-Ishii-Lions lemma; self-citations occur only in preliminary framework results and are not load-bearing for the main theorem.

full rationale

The paper's central claim, Theorem 3.9, is a comparison principle for the occupied PDE (3.12a). Its proof does not assume the conclusion: it constructs coercive approximations w_gamma and u_gamma, projects to finite dimensions via Lemma 4.7, applies the classical Crandall-Ishii-Lions lemma [13, Theorem 3.2] in Lemma 5.1, and proves the required CIL property of the Hamiltonian in Lemma 6.2 directly from Assumption 3.4. The target inequality u <= w is never used as an input; no parameter is fitted to the quantity being derived. Self-citations do appear: Lemma 3.2 imports OSDE well-posedness from the coauthor's thesis [36], and the occupation-derivative formalism is taken from [35]. These are framework facts with proofs in the cited prior work and are not the evidence supporting the comparison or uniqueness argument. The reader-flagged concern that Theorem 3.10 compares only bounded functions while Assumption 3.4 permits linear-growth payoffs is a genuine gap between the stated hypotheses and the conclusion, and the asserted infinite-dimensional limit in (6.5), cited to [13, Lemma 3.1], may need additional justification. These are correctness and technical-support issues, not circularity: the theorem does not reduce by construction to its own inputs or to an unverified self-citation chain.

Assumptions & free parameters 0 free parameters · 5 assumptions · 2 invented entities

No parameters are fitted to data: this is a pure mathematics paper. The constants c_*, T, and the separating family (f_k) are structural assumptions on the problem inputs, stated as Assumption 3.4 and Section 2.2; none are chosen post hoc to make the comparison principle hold. The main unproven inputs are prior results by the same authors ([35, 36]) and the classical finite-dimensional CIL lemma.

assumptions (5)
  • domain assumption Assumption 3.4: λ, b, σ, ℓ, g satisfy growth and Lipschitz conditions with constant c_* in the cylindrical norm, and λ ≥ 1/c_*.
    Invoked throughout: well-posedness of the OSDE (Lemma 3.2), finiteness of the objective J, the viscosity property of v, the CIL property (Lemma 6.2), and the limit passages in the comparison proof.
  • standard math There exists a separating family (f_k) ⊂ C¹_b(R^d) with Σ_k ||f_k||²_{C¹} ≤ 1 and f₀ constant, used to build the cylindrical norm and projections.
    Standard construction; the normalization and tail decay drive the K → ∞ passages in Lemma 5.1 and the ε ↓ 0 estimates in Section 6 Step 4.
  • domain assumption Existence and uniqueness of strong solutions of the controlled OSDE (3.2)-(3.3), cited from [36, Theorem 4.2.11].
    Used in Lemma 3.2 and in the regularity proof (Proposition 3.5, Lemma B.1); the cited source is the PhD thesis of a co-author.
  • standard math The Itô formula (3.6) for C^{1,2} functions of occupied processes, proved in Appendix A.
    Bridges the occupation derivative to the dynamics; the proof relies on the classical Itô formula and an approximation by piecewise-constant paths.
  • standard math Classical finite-dimensional Crandall-Ishii-Lions lemma [13, Theorem 3.2], applied on the closed set R^K × R^d including boundary points (Remark 5.2).
    The projection step of Lemma 5.1 rests on this; boundary jets are handled by the local jet definition in [13].
invented entities (2)
  • Occupation derivative ∂_o φ(o,x) = lim_{h↓0} (φ(o+hδ_x, x) - φ(o,x))/h
    purpose: Serves as the time-like derivative in the occupied PDE (1.3)/(3.12), replacing the classical time derivative in parabolic equations.
    Defined in (2.1); justified by the Itô formula (3.6) and the relation ∂_o φ = δ_o φ(x) in (2.3) for regular test functions. It is a mathematical construction, not an empirically postulated entity, so independent physical evidence is not applicable.
  • Cylindrical norm ρ(o,x) = (Σ_k |o(f_k)|² + |x|²)^{1/2}
    purpose: Penalization metric for the doubling-of-variables argument and for defining coercive approximations; makes the finite-dimensional projection method tractable.
    Constructed in Section 2.2 from the separating family; its metric properties on the non-compact M_T are asserted, not proved, which is flagged in red_flags.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Controlled Occupied Processes and Viscosity Solutions." pith.science (2026). https://pith.science/paper/N56NEY5E

@misc{pith2026241112080,
  author       = {Pith},
  title        = {Pith review of: Controlled Occupied Processes and Viscosity Solutions},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/N56NEY5E}},
  note         = {Machine review of arXiv:2411.12080}
}
abstract

We consider the optimal control of occupied processes which record all positions of the state process. Dynamic programming yields nonlinear equations on the space of positive measures. We develop the viscosity theory for this infinite dimensional parabolic $occupied$ PDE by proving a comparison result between sub and supersolutions, and thus provide a characterization of the value function as the unique viscosity solution. Toward this proof, an extension of the celebrated Crandall-Ishii-Lions (second order) Lemma to this setting, as well as finite-dimensional approximations, is established. Examples including the occupied heat equation, and pricing PDEs of financial derivatives contingent on the occupation measure are also discussed.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 2 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Viscosity Solutions of Fully second-order HJB Equations in the Wasserstein Space

    math.OC 2025-01 conditional novelty 7.0 of 10

    The value function of mean field control with common noise is the unique viscosity solution of a fully second-order HJB equation in the Wasserstein space.

  2. Cylindrical Projections of Occupied Diffusions

    math.NA 2026-04 unverdicted novelty 6.0 of 10

    Replacing the occupation measure by K cylindrical coordinates in a partition of unity gives strongly convergent (O(1/K)) finite-dimensional SDE approximations of occupied diffusions.

Reference graph

Works this paper leans on

40 extracted references · 33 canonical work pages · cited by 2 Pith papers

  1. [1]

    A vellaneda, A

    M. A vellaneda, A. Levy, and A. Parás. Pricing and hedging derivative securities in markets with uncertain volatilities. Applied Mathematical Finance , 2(2):73–88, 1995

    work page 1995

  2. [2]

    Bayraktar, I

    E. Bayraktar, I. Ekren, and X. Zhang. Comparison of visco sity solutions for a class of second order PDEs on the Wasserstein space. arXiv:2309.05040, 2023

    arXiv 2023

  3. [3]

    Bernard and Z

    C. Bernard and Z. Cui. Pricing timer options. Journal of Computational Finance , 15, 08 2010. 21 Viscosity Solutions Occupation Measures

    work page 2010

  4. [4]

    Bertucci and P

    C. Bertucci and P. L. Lions. An approximation of the squar ed Wasserstein distance and an application to Hamilton-Jacobi equations. arXiv:2409.11793, 2024

    arXiv 2024

  5. [5]

    Borwein and Q

    J. Borwein and Q. Zhu. Techniques of Variational Analysis . Springer, New York, NY, 2005

    work page 2005

  6. [6]

    Bouchard and X

    B. Bouchard and X. Tan. On the regularity of solutions of s ome linear parabolic path- dependent PDEs. Cahier de recherche CEREMADE, Université Paris Dauphine-P SL, 2023

    work page 2023

  7. [7]

    Burzoni, V

    M. Burzoni, V. Ignazio, A. M. Reppen, and H. M. Soner. Visc osity solutions for controlled McKean–Vlasov jump-diffusions. SIAM Journal on Control and Optimization , 58(3):1676– 1699, 2020

    work page 2020

  8. [8]

    Brownian particles controlled by their occupation measure

    L. Béthencourt, R. Catellier, and E. Tanré. Brownian par ticles controlled by their occupation measure. arXiv:2404.06960, 2024

    work page Pith review arXiv 2024

Show all 40 references
  1. [9]

    Cardaliaguet

    P. Cardaliaguet. Weak solutions for first order mean-field games with local cou pling. Analysis and Geometry in Control Theory and its Applications. , volume 11. Springer INdAM Series, 2015

    2015

  2. [10]

    Cardaliaguet, F

    P. Cardaliaguet, F. Delarue, J.-M. Lasry, and P.-L. Lio ns. The Master Equation and the Convergence Problem in Mean Field Games:(AMS-201) , volume 201. Princeton University Press, 2019

    2019

  3. [11]

    Carmona and F

    R. Carmona and F. Delarue. Probabilistic Theory of Mean Field Games with Applications I-II. Probability Theory and Stochastic Modelling. Springer In ternational Publishing, 2018

    2018

  4. [12]

    Cosso, F

    A. Cosso, F. Gozzi, I. Kharroubi, H. Pham, and M. Rosesto lato. Master Bellman equation in the Wasserstein space: Uniqueness of viscosity solution s. Transactions of the American Mathematical Society, 377(01):31–83, 2024

    2024

  5. [13]

    Crandall, H

    M. Crandall, H. Ishii, and P.-L. Lions. User’s guide to v iscosity solutions of second order artial differential equations. Bulletin of the American Mathematical Society , 27, 07 1992

    1992

  6. [14]

    M. G. Crandall and H. Ishii. The maximum principle for se micontinuous functions. Differ- ential and Integral Equations , 3(6):1001 – 1014, 1990

    1990

  7. [15]

    B. Dupire. Functional Itô calculus. SSRN, 2009. Republished in Quantitative Finance, 19(5):721–729, 2019

    2009

  8. [16]

    I. Ekeland. On the variational principle. Journal of Mathematical Analysis and Applications , 47(2):324–353, 1974

    1974

  9. [17]

    Ekren, C

    I. Ekren, C. Keller, N. Touzi, and J. Zhang. On viscosity solutions of path dependent PDEs. The Annals of Probability , 42(1):204 – 236, 2014

    2014

  10. [18]

    Ekren, N

    I. Ekren, N. Touzi, and J. Zhang. Viscosity solutions of fully nonlinear parabolic path dependent PDEs: Part I. The Annals of Probability , 44(2):1212 – 1253, 2016

    2016

  11. [19]

    Ekren, N

    I. Ekren, N. Touzi, and J. Zhang. Viscosity solutions of fully nonlinear parabolic path dependent PDEs: Part II. The Annals of Probability , 44(4):2507 – 2553, 2016

    2016

  12. [20]

    Fabbri, F

    G. Fabbri, F. Gozzi, and A. Swiech. Stochastic optimal c ontrol in infinite dimension. Prob- ability and Stochastic Modelling. Springer , 2017

    2017

  13. [21]

    W. H. Fleming and H. M. Soner. Controlled Markov processes and viscosity solutions , vol- ume 25. Springer Science & Business Media, 2006

    2006

  14. [22]

    Guyon and P

    J. Guyon and P. Henry-Labordère. Nonlinear Option Pricing . Chapman and Hall/CRC Financial Mathematics Series. CRC Press, 2013

    2013

  15. [23]

    Göllmann, D

    L. Göllmann, D. Kern, and H. Maurer. Optimal control pro blems with delays in state and control variables subject to mixed control–state constrai nts. Optimal Control Applications and Methods, 30:341 – 365, 07 2009

    2009

  16. [24]

    Lasry and P.-L

    J.-M. Lasry and P.-L. Lions. A remark on regularization in Hilbert spaces. Israel Journal of Mathematics , 55:257–266, 1986. 22 Viscosity Solutions Occupation Measures

    1986

  17. [25]

    Lasry and P.-L

    J.-M. Lasry and P.-L. Lions. Mean field games. Japanese Journal of Mathematics , 2:229–260, 03 2007

    2007

  18. [26]

    P.-L. Lions. Viscosity solutions of fully nonlinear se cond-order equations and optimal stochas- tic control in infinite dimensions. III. Uniqueness of visco sity solutions for general second- order equations. Journal of Functional Analysis , 86(1):1–18, 1989

    1989

  19. [27]

    Lions and P

    P.-L. Lions and P. Souganidis. Homogenization of the ba ckward-forward meanfield games systems in periodic environments. Rendiconti Lincei - Matematica e Applicazioni , 31:733– 755, 02 2021. doi: 10.4171/RLM/912

    2021 doi

  20. [28]

    A. Ocello. Controlled superprocesses and HJB equation in the space of finite measures. arXiv:2306.15962, 2023

    2023 arXiv

  21. [29]

    Peng and F

    S. Peng and F. Wang. BSDE, path-dependent PDE and nonlin ear Feynman-Kac formula. Science China Mathematics , 08 2011

    2011

  22. [30]

    Ren and F.-Y

    P. Ren and F.-Y. Wang. Derivative formulas in measure on Riemannian manifolds. Bulletin of the London Mathematical Society , 53(6):1786–1800, 2021

    2021

  23. [31]

    Revuz and M

    D. Revuz and M. Yor. Continuous Martingales and Brownian Motion . Springer Berlin, Heidelberg, 1999

    1999

  24. [32]

    Y. F. Saporito and J. Zhang. Stochastic control with del ayed information and related non- linear master equation. SIAM Journal on Control and Optimization , 57(1):693–717, 2019

    2019

  25. [33]

    H. M. Soner and Q. Yan. Viscosity solutions for Mckean–V lasov control on a torus. SIAM Journal on Control and Optimization , 62(2):903–923, 2024

    2024

  26. [34]

    H. M. Soner and Q. Yan. Viscosity solutions of the Eikona l equation on the Wasserstein space. Applied Mathematics and Optimization , 90(1), 2024

    2024

  27. [35]

    Tissot-Daguette

    V. Tissot-Daguette. Occupied processes: Going with th e flow. arXiv:2311.07936, 2023

    2023 arXiv

  28. [36]

    Tissot-Daguette

    V. Tissot-Daguette. Free boundaries, functional expa nsions, and occupied processes. PhD Thesis, Princeton University , 2024. A vailable at ProQuest Dissertations & Theses Global

    2024

  29. [37]

    J. Zhang. Backward Stochastic Differential Equations: From Linear to Fully Nonlinear Theory. Probability Theory and Stochastic Modelling. Springer Ne w York, 2017

    2017

  30. [38]

    J. Zhou. Viscosity solutions to second order path-depe ndent Hamilton-Jacobi-Bellman equa- tions and applications. Ann. Appl. Probab. , 33(6B):5564–5612, 2023

    2023

  31. [39]

    J. Zhou, N. Touzi, and J. Zhang. Viscosity solutions for HJB equations on the process space: Application to mean field control with common noise. arXiv:2401.04920, 2024

    2024 arXiv

  32. [40]

    Święch and L

    A. Święch and L. Wessels. Finite dimensional projectio ns of HJB equations in the Wasser- stein space. arXiv:2408.07688, 2024. 23

    2024 arXiv

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.