- Rating
- 2
- Confidence
- 3
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 2
Read review
Paper summary
This paper studies active learning for conditional flow matching in engineering design, where obtaining condition labels can be substantially more expensive than collecting candidate designs. The authors develop an analytical perspective based on a closed-form flow-matching formulation and a continuous piecewise-linear (CPWL) assumption. From this analysis, they argue that samples with similar conditions mainly improve generative diversity, whereas samples covering distant regions of the condition space improve conditional accuracy. Based on this observation, they propose two query strategies, and , respectively targeting diversity and accuracy, together with a weighted hybrid strategy. The approach is evaluated on one synthetic dataset and three engineering-design datasets using multiple active-learning baselines, 20 random seeds, supplementary statistical tests, proxy-error analyses, and experiments that relax the theoretical assumptions.
Critical strengths
- Rather than using generative models to assist conventional active learning, the work explicitly studies active learning for conditional generative models.
- The evaluation covers four datasets, multiple classical and recent active-learning baselines, repeated experiments over 20 random seeds, statistical comparisons, computational-cost analysis, proxy-error analysis, and tests under violations of the CPWL and closed-form assumptions.
Critical weaknesses
- The theoretical connection between the CPWL assumption and the proposed condition-space interpolation appears insufficiently justified.
- The practical query mechanism depends heavily on proxy-predicted labels for the unlabeled pool, which weakens the claimed active-learning interpretation.
Questions for authors
- Clarification: Can the authors provide a formal justification for Eq. (2), specifically explaining under what conditions and the vertices are guaranteed to lie in the same affine region of the conditional network for fixed ? How is the simplex subdivision of the condition space related to the actual activation-region partition of the trained CPWL network? If this cannot be guaranteed, which theoretical conclusions used to motivate and remain valid?
- Clarification: How sensitive are the final active-learning gains to the quality of the label proxy itself? In particular, could the authors compare using several proxy models with controlled prediction accuracy, and report performance as a function of proxy error? This would help determine whether the gains originate from the proposed acquisition principle or from the quality of the surrogate label predictor.
- Clarification: For the hybrid strategy , are the two scores normalized before linear combination? Since their individual terms involve distances and entropy changes with potentially different numerical scales, how stable is the meaning of across datasets and acquisition rounds?
The problem is interesting and the experimental effort is substantial, but the main theoretical argument currently relies on assumptions whose connection to the actual conditional network is not sufficiently established, while the practical method depends strongly on surrogate labels that are unavailable by definition in active learning. These issues make me lean against acceptance, although the paper is close enough that stronger theoretical clarification and a more convincing proxy-dependence analysis could materially change my assessment.
6 comments
To examine the approximation empirically, we have expanded Appendix C with quantitative comparisons in both single-simplex and multiple-simplex settings. The evaluation uses 1,000 Gaussian noise draws with corresponding target labels. Predictor outputs are compared at identical using the same noise parameterization. For generated endpoints, we compute each sample’s minimum MSE to its theoretical candidate set and then average over generated samples.
In the original two-condition Airfoil setting, labels 0 and 1 define a single one-dimensional simplex, with multiple training data points at each endpoint condition. The average nearest-candidate MSE is on the order of . Both predictor-output MSE and nearest-candidate MSE are largest near the target label farthest from the observed training labels, near the midpoint between the two endpoint conditions.
The multiple-simplex evaluation uses a conditional flow matching model trained on real data with their observed condition labels. We compare this trained model with two analytical references. The first uses the observed vertex conditions. The second groups multiple samples with nearby labels around each vertex, approximating nearby conditions by a shared vertex label. The second construction gives a smaller overall discrepancy between the analytical and trained predictor outputs, providing a closer empirical approximation to the trained model.
We use the paper’s time convention throughout: corresponds to Gaussian noise and to data. In the multiple-simplex comparisons, predictor MSE is approximately – at smaller flow times and increases overall as approaches one. This trend agrees with the unconditional closed-form comparison in Appendix H.2 under the same time convention. Along the label axis, both error measures exhibit multiple local peaks.
In the label-wise comparisons, average nearest-candidate MSE is numerically of a lower order of magnitude than predictor-output MSE. These measurements quantify agreement at the predictor and generated-endpoint levels, respectively.
2. Dependence on proxy quality and its relationship to acquisition performance
For unqueried candidates, our acquisition rules use performance labels predicted by the current proxy, and the selected samples receive true labels through querying. Predictive models and model-based estimates also guide querying in other active learning methods, including Committee, UEADA, ISP, and EMGM. Even when labels are not predicted explicitly, query decisions may depend on learned feature representations, as in CoreSet and ALIM. Our method uses predicted performance labels to evaluate its condition-space acquisition criteria.
We have expanded the proxy study in Appendix G to include RBF, Random Forest, SVR, and XGBoost, reporting proxy prediction MSE alongside downstream conditional error and diversity at acquisition rounds 1, 3, and 5.
“Random” denotes randomly sampled proxy hyperparameter configurations. “Large proxy error” denotes the ten configurations with the highest proxy MSE among 300 randomly sampled configurations. These high-error configurations provide an offline sensitivity diagnostic.
The results reveal a stage-dependent relationship. For each proxy family, prediction errors are relatively large in round 1, with dispersed downstream performance and a variable association between proxy MSE and the downstream metrics.
In rounds 3 and 5, the relationship within each proxy family becomes clearer: larger proxy MSE is generally associated with worse downstream performance—higher true conditional MSE for and lower generated-sample diversity for .
We report the results separately by proxy family and acquisition round. Within-family comparisons characterize sensitivity to prediction error, while comparisons across families examine variation associated with the proxy model. Prediction accuracy and local error structure can influence candidate rankings. The downstream results therefore reflect the acquisition rule together with the quality of the proxy information used to implement it.
3. Normalization of hybrid scores and interpretation of the hybrid weight
At each acquisition round, the complete scalar scores and for and are separately min–max normalized over the current input pool before combination. Section 2.4 documents this implementation detail alongside Eq. (7) and distinguishes acquisition scores from selected queries.
Larger places greater relative weight on the diversity-oriented score, while smaller emphasizes the accuracy-oriented score. The weight-sweep results in Section 3.2 and Appendix D.4 show dataset-dependent changes in the resulting metrics.
Accordingly, specifies the direction of the acquisition preference rather than quantitatively controlling the achieved diversity–accuracy trade-off. Its effect depends on the score distributions in the current candidate pool and the dataset. Section 4 identifies hybrid strategies providing quantitatively calibrated control of this trade-off as a direction for future work.
We thank the reviewer for the constructive comments. We have clarified the theoretical assumptions, expanded the quantitative analysis of the trained model and proxy dependence, and documented the normalization and interpretation of the hybrid strategy.
1. Conditions underlying Eq. (2) and the scope of the theoretical conclusions
Section 2.2 now explicitly states the condition-wise interpolation assumption and its sufficient condition. Appendix C discusses the condensation literature as background for data-dependent interpolation and provides quantitative comparisons between the analytical construction and the trained model.
For fixed (x′,t), a sufficient condition for Eq. (2) is that the entire condition simplex
C=convc0,…,cd
is contained in a single affine region of the map c↦ut(x′,c). Then, for
c∗=∑i=0daici,ai≥0,∑i=0dai=1,
we have
ut(x′,c∗)=∑i=0daiut(x′,ci).
For the trajectory-level analysis, this condition is assumed at the relevant state–time pairs along the sampling trajectory.
CPWL describes local affine behavior. Alignment between label-defined simplices and condition-affine regions is an additional analytical assumption. Our analytical model combines closed-form vertex predictors with a piecewise-affine extension in the condition variable:
ut(x′,c∗)=∑i=0daiutCF(x′,ci).
Here, CF denotes “closed form”: utCF is the closed-form noise predictor in Eq. (1). We retain ut for the trained network output and use the overbar to distinguish the ideal interpolated predictor.
The interpolation identity holds by construction for this analytical model. A trained network realizes this construction when it is affine in the condition variable within the simplex and matches the closed-form vertex outputs.
Appendix B now states the assumptions and arguments for the subsequent conclusions. Let Si denote the finite nonempty set of training data associated with condition ci. With positive normalized empirical sample weights, fixed barycentric coefficients, and standard Gaussian initialization, the ideal weighted flow converges almost surely to an endpoint in
\mathcal{S}(\boldsymbol{c}^{\ast})=\left\\{\sum_{i=0}^{d}a_i\boldsymbol{s}_i\\;\middle|\\;\boldsymbol{s}_i\in\mathcal{S}_i\right\\}.
This is the candidate set described in Eq. (3). The revised proof directly analyzes the weighted flow, including its convergence and candidate-endpoint characterization.
For Lemma B.2, we clarify the original inverse-function assumption by considering a C2 local right-inverse branch g, satisfying f(g(c))=c, with the participating vertex samples belonging to this same branch. Under the stated Lipschitz and second-derivative bounds, the revised conditional-error bound is
\left\\|f(\boldsymbol{x}^{\ast})-\boldsymbol{c}^{\ast}\right\\|_2\leq L_f\eta+K\max_{0\leq i,j\leq d}\left\\|\boldsymbol{c}_i-\boldsymbol{c}_j\right\\|_2^2,
where
\eta=\left\\|\boldsymbol{x}^{\ast}-\sum_{i=0}^{d}a_i g(\boldsymbol{c}_i)\right\\|_2.
We retain the original constants Lf, M, and K: Lf is the Lipschitz constant of f, M bounds the vector-valued second derivative of the local branch g, and K=Lf(d+1)M/2. The term Lfη accounts for deviation from the same-branch interpolant; when η=0, the bound reduces to the original quadratic form.
The candidate-set analysis provides the diversity upper-bound argument motivating QD, while the local interpolation-error bound motivates QA’s condition-coverage objective. These analytical conclusions apply under the stated interpolation and regularity conditions. The acquisition scores are computed from the data and proxy-predicted labels, and their empirical performance is assessed through the solver-based evaluations.