- Rating
- 2
- Confidence
- 3
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 2
Read review
Paper summary
This paper introduces PKNO, a Koopman neural operator for PDEs whose physical conditions, such as forcing, change over time. A shared encoder maps the recent field history to fixed latent coordinates, and a network generates a linear propagator for each Fourier mode from the current condition and a summary of the predicted history. On Navier-Stokes with a time-varying forcing amplitude, PKNO has lower rollout error than conditional FNO and KNO, while on shallow water it is worse than conditional FNO.
Critical strengths
The motivation is clear: many physical systems change their dynamics over time, and most neural operators do not model this change explicitly.
Critical weaknesses
The test of time-varying conditions is limited. The only changing condition is a scalar amplitude in [0.7, 1.3] that scales a fixed forcing pattern, there is no test outside the training range, and on shallow water PKNO has about twice the error of conditional FNO (0.0416 vs. 0.0197, Table 1).
The main design idea is not what is actually used. The paper motivates PKNO by multiplying linear propagators in fixed latent coordinates (Eq. 6), but in practice each propagator also depends on the predicted history, and the history is encoded again at every step (Appendix A.4), so the model works as a standard step-by-step predictor.
The ablation does not show where the gain comes from. The history summary is not removed in Table 2, so the improvement over conditional KNO may come from this extra input rather than from the proposed observables or the conditioned propagator.
The comparison is not strong enough. Only two conditional baselines are used, one of which is built by the authors, closely related methods such as PK-NN (Guo et al., 2025) and CAPE (Takamoto et al., 2023) are not compared, and all results use a single random seed.
The theoretical results add little. Theorem 1 follows from compactness for any continuous map and gives no bound on the dictionary size, and Theorem 2 is a standard error bound whose constant is not measured.
Some results are difficult to interpret. Burgers is a single-step task without rollout, and the fRMSE values in Table 1 (e.g., 2850 for IKNO) are not consistent with the normalized definition in Eq. 13.
Questions for authors
What is the error when the history is encoded once and only the conditioned propagators are applied, as in Eq. 6?
How does PKNO perform without the history summary, and how does conditional KNO perform when it receives the same summary?
How does PKNO perform when the forcing amplitude is outside the training range, or when the forcing pattern or viscosity changes over time?
Why are many fRMSE values in Table 1 much larger than 1?
9 comments
Response to Reviewer 9xRf (Part 2 of 4)
W2: The main design idea ... & Q1: What is the error ...
Encoding the initial history once and composing the condition-dependent propagators as in Eq. (6) yields a Full of 0.03585 on the 40-step time-varying Navier–Stokes rollout () and 0.0160 on fixed Navier–Stokes. The corresponding errors for complete PKNO (M6) are 0.01581 and 0.0150. See Tables 1 and 2 in our response to AC for the propagation variants and full metrics, and Table 3 below for the accompanying comparison.
(i) Re-encoding improves rollout accuracy. M1 initializes and composes . Adding step-wise re-encoding while keeping the condition-only propagator (M3) reduces time-varying Full from 0.03585 to 0.01809 (49.5%). Adding latent-state adaptation without re-encoding (M2) reduces it to 0.02785 (22.3%).
(ii) History adaptation adds a gain at matched re-encoding. With identical re-encoding, adding the rolling history summary (M3 to M6) lowers time-varying Full from 0.01809 to 0.01581 (12.6%) and from 0.001446 to 0.001305 (9.7%). On fixed Navier–Stokes, Full decreases from 0.0154 to 0.0150.
(iii) Mode-wise propagation improves on the matched nonlinear update. M5 replaces the Koopman matrix with a parameter-matched nonlinear latent update while keeping the encoder, decoder, inputs and step-wise re-encoding identical. Relative to M5, M6 lowers Full by 12.2% (0.01581 vs. 0.0180) and by 15.8% (0.001305 vs. 0.001550).
Table 3. Rollout and high-frequency errors for the Navier–Stokes ablations ().
| Condition | Variant | Full ↓ | ↓ |
|---|---|---|---|
| Fixed NS | M1 | 0.0160 | 0.0002 |
M3 | 0.0154 | 0.0001 | |
M6 (PKNO) | 0.0150 | 0.0001 | |
| Time-varying NS | M1 | 0.0359 | 0.0014 |
M2 | 0.0279 | 0.0012 | |
M3 | 0.0181 | 0.0014 | |
M5 | 0.0180 | 0.0016 | |
M6 (PKNO) | 0.0158 | 0.0013 | |
| Cond. KNO | 0.0248 | 0.0021 | |
Cond. KNO+H (M7) | 0.0185 | 0.0012 |
The practical closed-loop rollout decomposes exactly into the propagator product of Eq. (6) plus step-wise re-encoding corrections. Defining the one-step re-encoding residual as with , each closed-loop latent update satisfies . Expanding this recursion over steps yields
The first term is the ordered product of the history-adaptive propagators along the realized trajectory, in the shared coordinates of Eq. (6). The second term accounts for the accumulated step-wise re-encoding corrections. We will include the propagation ablations in Section 4.2, incorporate this decomposition as Proposition 1 in Section 3.3, and clarify in Section 3.2 that Eq. (6) defines the shared-coordinate composition core while Section 3.3 describes its closed-loop realization.
TO CONTINUE
Response to Reviewer 9xRf (Part 3 of 4)
W3: The ablation ... & Q2: How does PKNO perform ...
As shown in Table 3, on time-varying Navier–Stokes (), dropping the explicit history summary from PKNO (M3) gives a Full error of 0.01809, which is still 27.1% lower than standard Cond. KNO (0.02480). Conversely, giving Cond. KNO a matched history-summary module (M7) reduces its error to 0.01852, and full PKNO (M6) lowers this by another 14.6% to 0.01581. Thus, while the history summary accounts for part of the improvement, PKNO still holds a clear rollout advantage when both models use the same history input.
Controlled ablations disentangle the explicit history summary from the hybrid observables and condition-dependent propagator. In PKNO, the rolling field history enters through two pathways: the shared observable encoder and the context summary supplied to the propagator generator. Table 3 gives the matched comparisons under the same 40-step rollout protocol. PKNO w/o H (M3) drops from the generator so that while retaining hybrid observables and step-wise re-encoding. Cond. KNO+H (M7) keeps the standard Cond. KNO dictionary and decoder but conditions its propagator on both and an independently trained history summary matching PKNO's architecture, dimension, input history, and update frequency ().
(i) PKNO retains gains without the summary. Compared with Cond. KNO, M3 combines hybrid observables with condition-dependent propagation and reduces Full from 0.02480 to 0.01809 (27.1%). It reduces from to (31.9%).
(ii) Matched history inputs preserve the rollout gain. Adding within PKNO (M3 to M6) lowers Full from 0.01809 to 0.01581 (12.6%). Adding condition and history-summary inputs jointly to the KNO baseline (M7) lowers its Full from 0.02480 to 0.01852 (25.3%). With matched inputs, M6 has 14.6% lower Full than M7 (0.01581 vs. 0.01852). M7 has lower ( vs. for M6). Together with Table 2 in the main paper, the comparisons show that the explicit history summary accounts for part of the gain.
Hybrid observables and history-conditioned propagation act through distinct mathematical pathways in our theoretical framework. Theorem 1 shows that a condition-independent dictionary and a continuous coefficient map can uniformly approximate one-step target observables (), motivating our separation between shared hybrid observables and condition-dependent propagation. For the closed-loop predictor , taking the Fréchet derivative with respect to a history perturbation (with ) separates two distinct mechanisms:
Accordingly, in the rollout error bound of Theorem 2 (), the sufficient Lipschitz constant decomposes as with . Removing in M3 zeroes out the second pathway () while preserving the shared observable pathway (). We will incorporate the M3 and M7 comparisons and this pathway discussion into Section 4.2 and Appendix B of the revised manuscript.
W4: The comparison is not strong enough ...
We added PK-NN and CAPE and used the same hyperparameter-search budget for each model. Table 2 in our response to Reviewer mDV2 reports mean standard deviation over three seeds.
TO CONTINUE
Response to Reviewer 9xRf (Part 1 of 4)
Dear Reviewer 9xRf,
We sincerely appreciate your thoughtful and constructive comments. We have carefully considered each of your questions and provide detailed responses below. All changes are in green in the revised PDF.
The large fRMSE values in the original Table 1 arise from normalization by the target energy in the high-frequency band, which can be nearly zero for smooth solutions. The fixed-condition Burgers benchmark evaluates one-step prediction across initial profiles, while multi-step results for Navier–Stokes and shallow water are reported in Table 6 in our response to AC and Table 2 in our response to Reviewer mDV2.
Theoretically, let Us,t(k) and Us,t(k) be the Fourier coefficients of the target and prediction for sample s at time t. We have
fRMSE=∑s,t∑k∈H∣Us,t(k)∣2∑s,t∑k∈H∣Us,t(k)−Us,t(k)∣2,
where $H=\{\mathbf{k}\in\mathcal K: \|\mathbf{k}\|_2/k_{\max}\geq 2/3\}$, with $k_{\max}=\max_{\mathbf{k}\in\mathcal K} \|\mathbf{k}\|_2$. Our numerical audit confirms that the target energy in this band can be extremely small, accounting for only 3.040×10−16 of the total energy on fixed Burgers. Consequently, even small high-frequency prediction errors can produce large fRMSE values.
In the tables below, we normalize the high-frequency error by the total target energy, giving
RHF=∑s,t∑k∣Us,t(k)∣2∑s,t∑k∈H∣Us,t(k)−Us,t(k)∣2.
We will replace fRMSE with RHF in Table 1 and update Eq. (13) in the revised manuscript.
PKNO achieves the lowest Full L2 among the evaluated methods when the forcing amplitude exceeds the training range, when the spatial forcing pattern varies, and when viscosity varies over time. See Table 7 in our response to AC for the data-generation settings.
Amplitude extrapolation and faster forcing paths. On Navier–Stokes at ν=10−3, we train on forcing amplitudes qn∈[0.7,1.3]. Table 4 in our response to AC reports the tests for qn∈(1.3,1.5] and for faster forcing paths with 16 control knots. PKNO's Full L2 is 0.0180 in the extrapolation test and 0.0168 on the faster paths, compared with 0.0189 and 0.0180 for CAPE, the strongest baseline in both settings.
Time-varying spatial forcing. We replace the fixed profile q(t)f0(x,y) with fn(x,y)=∑j=18aj,nϕj(x,y), where the coefficients aj,n∈[−0.5,0.5] vary over time and the sine and cosine bases use wavevectors (1,1), (1,2), (2,1), and (2,2). Table 3 in our response to AC gives the full comparison. On held-out spatial profiles, PKNO reduces Full L2 by 19.7% relative to Cond. FNO and 30.0% relative to PKNO-scalar, which receives only ∣an∣2 under matched-norm conditions (Table 1).
Table 1. Rollout and high-frequency errors under time-varying spatial forcing (ν=10−3).
Time-varying viscosity. The Fourier diffusion term νnΔω(k)=−νn∥k∥22ωn(k) motivates mode-wise conditioning, since changing viscosity affects different frequencies by different amounts. We test νn∈[0.7,1.3]×10−3 with either fixed forcing amplitude q=1 or independently varying νn and qn. PKNO lowers Full L2 by 13.8% relative to Cond. FNO and 21.7% relative to Cond. KNO in the viscosity-only case, and by 22.9% and 29.3% under joint variation (Table 2).
Table 2. Rollout and high-frequency errors under time-varying viscosity.
Under the updated experimental settings and a consistent training budget across all methods, PKNO achieves a Full L2 of 0.01944 on shallow water, outperforming four baselines. The full comparison is reported in Table 6 in our response to AC. We will add these experiments and protocol details to Sections 4.1–4.2 and Appendix A.
TO CONTINUE