- Rating
- 3(+1)
- Confidence
- 2
- Informativeness
- 2
- Presentation
- 3
- Soundness
- 2
Read review
Paper summary
This paper introduces a quantum-inspired classical method for ranking node responses to a perturbation at a seed node. It trains a phased Szegedy walk while preserving unitary propagation and computes scores through a geometrically discounted coherent sum of walk amplitudes followed by a squared-magnitude readout. Learned parameters are attached to individual edges and nodes, allowing direct inspection of the fitted model. Evaluation covers six power grids and five Perturb-seq datasets, with perturbation seeds held out during training. Their model showcases improved performance on the trained random-walk baseline and the neural baselines on gene perturbations.
Critical strengths
-
A conceptually interesting step of including wave phases. I find the combination of a trainable Szegedy walk, a coherent-resolvent readout, and explicit node and edge parameters appealing. It allows the model to learn how propagation paths reinforce or cancel one another while keeping the fitted parameters directly inspectable.
-
Meaningful empirical gains supported by targeted ablations. Across five Perturb-seq datasets with held-out perturbation seeds, VQIL achieves mean Spearman correlation of 0.620, compared with 0.582 for GCN and 0.570 for trained RWR (Table 2). Ablations further support its design choices: adding node phases improves correlation from 0.587 to 0.620, while coherent geometric accumulation achieves 0.620 versus 0.517 for its square-then-add counterpart (Tables 2–3). Together, these results demonstrate predictive improvements over trained competitors and support the usefulness of node phases and coherent accumulation within the proposed architecture, although their advantage over comparably expressive alternatives remains unresolved.
Critical weaknesses
-
The conceptual novelty appears limited. VQIL combines an existing phased Szegedy construction with end-to-end ranking supervision and a coherent-resolvent readout. Phased walks, trainable quantum-walk operators, and graph-associated learned propagation parameters have precedents, as acknowledged in Section 3. The contribution therefore appears to lie primarily in the particular combination of parameterization, training objective, and readout, rather than introducing interference or learnable walk dynamics. I consider this combination potentially useful, but do not see a substantial conceptual advance without significant empirical results. This makes convincing evidence of an empirical or interpretability advantage over closely matched existing approaches particularly important; the current comparisons do not yet establish that advantage conclusively.
-
The comparisons do not sufficiently isolate the advantage of phase-based propagation. Interference and cancellation also occur in classical waves and signed models. While the Szegedy construction justifies "quantum-inspired," the novelty should concern the trainable formulation and coherent readout, rather than interference itself. The sheaf baseline uses feature-dependent, MLP-parameterized rotations, so it is not a closely matched, directly graph-parameterized control. Moreover, trained RWR learns edge weights with a fixed restart probability, whereas VQIL additionally learns node phases. A classical model with learned edge weights and per-node restart probabilities, alongside a matched cancellation-capable propagator, would help isolate the contribution. The ablations show that phases and coherent accumulation help within VQIL, but do not establish their advantage over comparably expressive alternatives.
-
Parameter visibility is better supported than mechanistic explainability. The parameter–annotation associations are useful, particularly the generator-bus result, but edge-weight correlations with co-expression are almost identical for VQIL and RWR (0.11–0.28 versus 0.11–0.27). The node-phase interpretation is not compared against a classical model with learned node parameters. Moreocer the paper does not explain which paths reinforce or cancel for individual predictions, or validate Section 6's stronger claim that phase magnitudes identify explicit sinks where perturbations are halted. Prediction-specific analyses or targeted intervention tests would strengthen the mechanistic interpretation.
-
The symmetric link phases appear mathematically redundant. In Eqs. (1)–(3), define . Since , commutes with the swap , giving , where has zero link phases but unchanged weights and node phases. The seed and readout states transform by the same , implying Under the equations as written, link phases therefore cannot affect predictions: their lack of empirical benefit reflects an exact redundancy rather than a dataset-dependent observation. This does not invalidate the main vertex-phased model, but requires clarification of the trainable parameter space.
Questions for authors
-
What evidence separates the benefit of the proposed mechanism from that of adding node-specific parameters? Are matched results available for RWR with learned edge weights and per-node restart probabilities, or a directly graph-parameterized signed/rotation-based propagator? Without such controls, how narrowly should the claimed advantage be interpreted?
-
Beyond parameter–annotation correlations, what supports Section 6's interpretation of as identifying sinks where perturbations are halted? Does the recurrent-response association persist when calculated on held-out perturbations rather than the training perturbations used in Figure 6?
-
Could you provide a parameter-matched comparison with a classical signed or rotation-based propagator, using the same number of learnable edge and node parameters, a comparable propagation horizon and readout, and the same training and evaluation protocol? This would help determine whether VQIL’s gains are specific to its propagation rule or could be reproduced by an equally parameterized classical model with node-level flexibility and cancellation capabilities.
-
Do Eqs. (1)–(3), with , imply , explaining the lack of link-phase benefit? Can you confirm that setting link phases to zero without retraining leaves predictions unchanged?
5 comments
Authors' Response (Part 2)
Intervention tests and "sinks". We intervened on the trained models without retraining. enters the readout through a factor on every walk passing through node (onward transmission, including its phase) and a factor on node 's own score (response gain).
| Spearman ρ | Phase angle | Voltage | PPI |
|---|---|---|---|
| Vertex-phased-VQIL (reference) | 0.9115 ± 0.0433 | 0.8306 ± 0.0564 | 0.6196 ± 0.1670 |
| δ = 0 at the 5% of nodes with largest abs(δ) | 0.8655 ± 0.0599 | 0.7777 ± 0.0718 | 0.5780 ± 0.1583 |
| δ = 0 at 5% random nodes | 0.9071 ± 0.0458 | 0.8204 ± 0.0605 | 0.6197 ± 0.1672 |
Mean ± sd over cells, untruncated readout.
High- nodes matter for the prediction: knocking them out costs 0.046–0.051, against 0.004–0.008 for random nodes. acts through the node's own response gain, not by blocking onward transmission. We therefore agree that "sinks where perturbations are halted" in Section 6 was imprecise, and we have aligned it with Section 5.2: tracks a node's local response propensity, low at voltage-regulated buses and high at recurrently responsive genes.
Classical model with learned node parameters. We trained RWR with a learned per-node restart probability (same number of parameters, same protocol). On the voltage response, identifies generator buses with AUC 0.7386 ± 0.1003, against 0.6398 ± 0.1111 for the per-node restart (30 cells; higher on 5 of 6 grids).
Edge weights. We agree: the VQIL and RWR weight fields carry the same co-expression information, so the readability of edge weights is shared with the classical walk. What VQIL adds is the node parameter. We now state this explicitly.
W2 & Q3 (parameter-matched classical propagator). We agree with the points of W2 and answer them in turn. All numbers refer to the table under Q1.
Sheaf baseline. We agree that Sheaf-O(2), with feature-dependent, MLP-parameterized rotations, is not a closely matched control. It remains in Tables 1–2 as a neural baseline only. The matched controls are (a)–(d) of Q1, which are parameterized directly on the graph: one scalar per edge and one per node, no features and no MLP.
Learned edge weights with per-node restart. This is control (a). It improves on Trained-RWR by / / (phase angle / voltage / PPI; 6/6 grids for each response, 5/5 PPI datasets), so a learned per-node restart accounts for part of the gap in Table 1. Vertex-phased-VQIL exceeds it in 27/30 voltage cells (5/6 grids) and in 25/25 PPI cells (5/5 datasets), and is level on phase angle (17/30 cells, 3/6 grids).
Matched cancellation-capable propagator. We trained both a signed and a rotation-based propagator, controls (b) and (c). They are classical, non-unitary propagators on the same graph with one learned angle per node:
where is the transition matrix of Trained-RWR and is the transmission of node : for the signed propagator and for the rotation-based one. A walk through nodes contributes with the factor , so walks reaching a node along different routes can cancel, by sign in (b) and by rotation in (c). Setting recovers Trained-RWR.
The comparison is matched to Vertex-phased-VQIL as follows.
- Parameters: edge weights with the parameterization of Eq. 2 and node angles, initialized and clamped exactly like (uniform in , clamped to ); the count is for all three models.
- Horizon: restart , the setting used for Trained-RWR in Tables 1–2.
- Readout: the signal is accumulated over walk lengths and the modulus is taken once at the end, as in the coherent readout of Eq. 3.
- Training and evaluation: identical folds, loss, optimizer, early stopping and model selection.
Cancellation helps the classical walk on the power grids: (b) and (c) improve on Trained-RWR by / and / (phase angle / voltage), and neither changes the PPI result. Vertex-phased-VQIL exceeds both in 30/30 phase-angle cells, 30/30 voltage cells and 25/25 PPI cells.
Advantage over comparably expressive alternatives. The ablations in the paper compare variants within VQIL only, as you note. The table under Q1 adds the comparison with classical models of the same parameter count (). At the setting of Tables 1–2, Vertex-phased-VQIL outperforms the matched signed and rotation-based propagators and the lazy walk in every grid and PPI dataset, and matches or exceeds the walk with learned per-node restart.
Authors' Response (Part 3)
W4 / Q4 (symmetric link phases).
Thank you for this precise observation. Your derivation is correct: for , the diagonal unitary commutes with the swap , so , and since the seed and readout states transform by the same ,
holds exactly. We also confirmed it numerically: setting in the trained Link-phased-VQIL models, without retraining, leaves all predictions unchanged up to floating-point precision (maximum deviation of order ).
This does not affect our results or conclusions. Our main model is Vertex-phased-VQIL, which has no link phases: its trainable parameters are the edge weights and the node phases , and all headline results, the readout ablation and the interpretability analysis use this variant. Link phases appear only in the Link-phased and Graph-phased ablation rows, which we included because the underlying graph-phased Szegedy walk admits them. Your argument explains exactly why those rows coincide with Standard-VQIL and Vertex-phased-VQIL (differences , due to random initialization only).
In the revision we state this invariance explicitly in Section 4.1, describe the Link-phased and Graph-phased rows as a consistency check, and remove link phases from the list of trained components in the abstract and contributions.
References
(a) Per-node restart W. Jin, et al. "Supervised and Extended Restart in Random Walks for Ranking and Link Prediction in Networks." PLOS ONE, 2019.
(b) Signed propagator Y. Tian and R. Lambiotte. "Spreading and Structural Balance on Signed Networks." SIAM Journal on Applied Dynamical Systems, 2024.
(c) Rotation-based propagator Y. Tian and R. Lambiotte. "Structural Balance and Random Walks on Complex Networks with Complex Weights." SIAM Journal on Mathematics of Data Science, 2024.
(d) Lazy walk P. Du, et al. "Supervised Lazy Random Walk for Topic-Focused Multi-Document Summarization." IEEE International Conference on Data Mining (ICDM), 2011.
We thank you for the detailed and constructive comments, which helped us isolate the role of the node phase with matched classical controls and sharpen the claims of the paper. We hope the responses above resolve your concerns.
Best regards,
Authors of Submission 61787
Authors' Response (Part 1)
We thank the reviewer for the detailed and technically precise review. We ran the controls you asked for and answer each point below.
W1& Q1 (empirical evidence against closely matched existing approaches). All controls you name are now available, together with a lazy walk. Each adds one learned scalar per node to Trained-RWR (m+n parameters, as in Vertex-phased-VQIL) and uses the protocol of Tables 1–2 (restart r=0.1, same steps, folds, loss and optimiser):
Full definitions are given under Q3.
Spearman ρ, mean ± sd over 30 cells (power grids) and 25 cells (PPI); m edges, n nodes.
The five folds of a task share one graph, so the unit of analysis is the grid (n=6) or the dataset (n=5). We report per-cell win counts together with the number of independent units won. Vertex-phased-VQIL exceeds (b) and (c) in 30/30 (phase angle) and 30/30 (voltage) grid cells and in 25/25 PPI cells, and exceeds (d) in 30/30, 30/30 and 24/25 cells. In all three comparisons it wins in all six grids for each response and in all five PPI datasets. Against (a) it wins in 27/30 voltage cells (5/6 grids) and in 25/25 PPI cells (5/5 datasets), and is level on phase angle (17/30 cells, 3/6 grids). None of the four classical walks matches VQIL on all three tasks at the same parameter count.
On the grids every control has a higher mean ρ than Trained-RWR, so node-specific parameters do account for part of the gap in Table 1. The remaining difference lies in how the node parameter enters the walk. The node phase is the phase θv=π+δv of the reflection at node v (Eq. 1) and acts inside a unitary step: it does not remove probability, hold the walker, or add moves outside the graph, and it changes the scores only through the coherent readout (Eq. 3), in which amplitudes are summed before the squared norm. The classical parameters instead remove or hold probability mass at the node, as in (a) and (d), or rescale or rotate the transmitted signal of a non-unitary propagator, as in (b) and (c). We therefore state the claim as follows: at the setting of Tables 1–2 and with the same number of parameters, Vertex-phased-VQIL outperforms weight-only RWR and controls (b)–(d) in every grid and PPI dataset, and matches or exceeds the per-node-restart walk (a) without any jump to the seed.
W3 & Q2 (support for the interpretation of |δ|).
Held-out perturbations. Yes, the recurrent-response association persists when it is computed only on the test perturbations of each fold, which were never used to fit δ. Recurrent response is the fraction of perturbations in which a gene is among the 10% strongest responders. We control for degree because degree is known from the graph without training and acts in the opposite direction: ∣δ∣ decreases with degree (ρ = −0.40 to −0.60), whereas responsiveness increases with it (ρ = 0.16 to 0.38).
Mean ± sd over five folds.