- Rating
- 3(+1)
- Confidence
- 3
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 2
Read review
Paper summary
The paper develops regression losses for conditioning diffusions with rank-deficient diffusion coefficients, including settings without smooth ambient-space densities. Using predictable representation in the state filtration, it identifies the minimum-energy Brownian control by regressing short projected or raw Brownian increments. A state-increment formulation recovers the induced drift and, via the pseudoinverse, the corresponding control. The analysis covers random initialization and density processes that may vanish, and establishes population convergence on compact intervals before the terminal time under the stated well-posedness, local absolute-continuity, square-integrability, and feedback assumptions. The paper also presents an endpoint-control formulation and illustrates the method on a nonlinear diffusion with a numerical Doob-bridge reference and a Fashion-MNIST inpainting proof of concept on a fixed affine subspace.
Critical strengths
- A regression formulation that accommodates degenerate noise. Theorems 1–2 and Corollary 1 turn predictable representation into regression targets that require neither an ambient-space score nor reconstruction of the entire driving Brownian motion. They explain which noise directions affect the state and how the raw-increment formulation avoids numerical projection. This broadens the setting relative to the Euclidean derivation using invertible volatility in Pidstrigach et al. (2025).
- Careful treatment of the assumptions behind the main results. Appendix C handles random initialization and changes of completed filtrations, and the localization argument accommodates density processes that may vanish. Theorem 7 accounts for initial entropy. The paper distinguishes existence of a controlled solution under a changed measure from weak uniqueness of independently simulated solutions. Appendix G also distinguishes convergence for individual conditions from pooled convergence and treats discretization of the projected and raw targets separately.
Critical weaknesses
- The image experiment needs a clearer link to the regression theory. Section 5.3 trains under the empirical bridge-mixture law using a finite-window increment. Theorem 11 establishes an exact identity under the additive reference law. Specifying how the implemented increments are constructed and deriving their conditional expectation under the actual training law would clarify which drift the regression identifies.
- The numerical evidence supports a narrower set of claims.Figure 1 gives a qualitative comparison with the numerical Doob bridge, but leaves the accuracy of the three increment windows unquantified. Figure 2 illustrates inpainting feasibility. Its stronger claims about conditional scores and time reversal need a separate argument in the 648 accessible coordinates, since the absence of an ambient-space density alone does not establish an obstruction on the accessible support.
Questions for authors
ould you clarify which admissibility conditions justify the entropy identity used for arbitrary competitors in the proof of Theorem 10? Theorem 7 explicitly includes its corresponding identity in the definition of admissibility. Does Theorem 10 require a similar condition? Please also clarify whether the uniqueness claim concerns the control, the state-path law, or the pair . 2. How are the increments in Section 5.3 constructed under , and what is their conditional expectation given ? Does the Gaussian bridge calculation give exact population identification for the implemented finite window? Please distinguish identification under the empirical mixture from generalization to held-out conditions. 3. Which obstruction to conditional scores or time reversal remains in the 648 accessible coordinates? Please distinguish ambient-space densities from densities on the accessible support. 4. Could you quantify the discrepancy from the numerical Doob bridge for the three increment windows, for example using effective-drift or endpoint error, and report the simulation time step and terminal cutoff? 5. Could the abstract replace “à la Üstünel” with a direct explanation of how predictable representation leads to the regression targets and specify the assumptions behind the population result? Please distinguish these from the additional assumptions used for discretization and learning in Appendix G. Section 2.1 also appears to omit the conditioning when it states that the original and controlled processes have the same law. 6. Could you label the ground truth, observed condition, and generated completion directly in Figure 2? If the figure is intended to support a comparative advantage, what evidence supports that claim?
2 comments
4. Could you quantify the discrepancy from the numerical Doob bridge for the three increment windows, for example using effective-drift or endpoint error, and report the simulation time step and terminal cutoff?
We have added further quantitative results that address this concern in Appendix . Our quantitative results also include a comparison for the different losses we have introduced.
5. Could the abstract replace “à la Üstünel” with a direct explanation of how predictable representation leads to the regression targets and specify the assumptions behind the population result?
We appreciate the suggestions. We believe the second part would be too technical to include in an abstract. We included the first part in the abstract in the updated manuscript.
Please distinguish these from the additional assumptions used for discretization and learning in Appendix G.
We have added a short discussion in the main body after introducing Algorithm~1.
Section 2.1 also appears to omit the conditioning when it states that the original and controlled processes have the same law.
Corrected, thanks.
6. Could you label the ground truth, observed condition, and generated completion directly in Figure 2? If the figure is intended to support a comparative advantage, what evidence supports that claim?
We have added labels to the figure.
The figure was to demonstrate that one can generate these images without relying on local (intrinsic) information, such as local coordinates under which a smooth conditional likelihood (with respect to Lebesgue measure) may or may not exist. We did not aim for any quantitative claim. We have added a brief clarification in the main body related to this. However, in the current version, there are several other numerical examples of the updated manuscript can be found in the Appendix H.
We thank to the reviewer for pointing out several missing details from the text and explicitly pointing them out. We believe we have addressed most of these concerns, and they can be seen on the updated version of the manuscript, marked with blue. Below, we point out how the issues brought up are addressed by referring to each one separately.
Clarification on Theorem 10
We have added the missing necessary finite entropy condition for the ``admissibility'' in Theorem 10 so that the relative entropy is well defined. The uniqueness claim concerns the pair (Q,v). This then implies the uniqueness of the control and the induced state-path law. We have highlighted the uniqueness of (Q,v) in the statement and added a remark. The proof remains the same.
How are the increments in Section 5.3 constructed under RνT, and what is their conditional expectation given (t,Xt,Y)?
We first sample a terminal target Z∼νT and set Y=G(Z). Writing L=AS, we jointly sample Bt and Bt+ε from the Gaussian bridge ending at L†(Z−x0).
We use the empirical estimator ERνT[εBt+ε−Bt∣t,Xt,Y], which satisfies $\mathbb{E}^{R_{\nu_T}}[\tfrac{B_{t+\varepsilon}-B_t}{\varepsilon}\mid t,X_t,Y] =\tfrac{\mathbb{E}^{R_{\nu_T}}[B_T-B_t\mid t,X_t,Y]}{T-t}$.
Please distinguish identification under the empirical mixture from generalization to held-out conditions
We included an explain under Figure 2.
Specifying how the implemented increments are constructed and deriving their conditional expectation under the actual training law would clarify which drift the regression identifies.
We have included a further discussion related to this implementation.
Does the Gaussian bridge calculation give exact population identification for the implemented finite window?
It does. This is a special case of Theorem~11 in the Appendix, so no ε→0 is required.
Please distinguish identification under the empirical mixture from generalization to held-out conditions.
We have highlighted this in a sentence.
Q3. Which obstruction to conditional scores or time reversal remains in the 648 accessible coordinates? Please distinguish ambient-space densities from densities on the accessible support.
The main ones are that, ∇xlogp(Y=y∣Xt=x) is still singular at t=T even under the 648 accessible coordinates, which is required to be well defined for the usual time-inversion. Furthermore, the usual conditional score method to learn ∇xlogp(Y=y∣Xt=x) is to learn both ∇xlogpt(x∣Y=y) and ∇xlogpt(x); however, since the initial condition is deterministic and the SDE is on 648 dimension, at t=0 we have singularity for ∇xlogpt(x), which is the marginal distribution of X at t=0, since the initial condition we have taken is deterministic.
There is no direct obstruction to conditional score under the 648 accessible coordinates in the region 0<t<T (however, since this is an open interval there will be large error). However, for most applications we do not know the actual accessible dimensions (and computing them can be expensive). Furthermore, the accessible dimensions might change for all t with positive probability, so in general one needs to understand all the singularities of the score function. In each of these cases, one needs to pass from the
extrinsic space'' (or the ambient space) to an intrinsic manifold to access to the full support. In most applications, this knowledge is not available directly, as the diffusion $X$ merely lies on anoisy manifold'', where the dimension of the support X can very with positive probability. We refer toJakiw Pidstrigach. Score-Based Generative Models Detect Manifolds. NeurIPS 2022,
and references therein for further discussion on this matter. In principle, the singularities of ∇logxlogpt and ∇xlogp(Y=y∣Xt=x) can also incur in the interior interval (0,T). Our aim was to demonstrate that one can do the conditioning without relying on any intrinsic information, which amounts to the 648 accessible coordinates in that example. We remark that, for generative models, extrinsic structure is always available. We have added further details on this in the caption of Figure 2.