- Rating
- 2
- Confidence
- 2
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 2
Read review
Paper summary
This paper aims to solve sensor-wise domain shift in unsupervised domain adaptation for wearable multivariate time series, where channel semantics change across users or wearing conditions. Specifically, the authors propose Inter-Domain Sensor Alignment (IDSA), which learns a source–target transport matrix to align target channels with the source sensor space, and a channel decorrelation regularizer to preserve distinguishable representations. Experiments on Opportunity, WISDM, NinaPro DB5, and SD-gesture demonstrate improved average classification accuracy and F1-score across four UDA backbones, together with stronger performance preservation under controlled sensor permutations than the evaluated baselines.
Critical strengths
-
The paper addresses a practically relevant obstacle to reliable wearable sensing.
-
The design is feasible and easy to follow.
-
The evaluation includes real public datasets spanning inertial sensing and sEMG, ten main UDA baselines, and additional OT comparisons.
Critical weaknesses
-
The diagnostic setup allows rotation and scaling, but IDSA’s mixing matrix is much more restrictive.
-
The OT proof does not show that joint training will find the correct sensor mapping.
-
The paper does not quantitatively test whether the sensor mapping is correct.
-
The validation setup is unclear, especially whether tuning uses target labels.
-
IDSA uses target data, while the foundation-model baselines do not.
Questions for authors
see above
6 comments
Response to Critical Weakness 2 (Uncertain correspondence objective).
Alternative cost analysis
We additionally investigated whether IDSA can utilize alternative cost functions for estimating sensor-wise relationships. Specifically, we considered four alternatives based on different sensor characteristics: eq6, which corresponds to the temporal-profile cost in Eq. (6); logpsd, which measures differences between log-transformed power spectra; pshape, which compares normalized spectral distributions using the Hellinger distance; and combo, which combines spectral information with the mean signal level. For the spectral-based costs, we computed the mean periodogram across samples after removing the temporal mean of each window.
We evaluated on the SD-gesture dataset by forcing permutation on the target domain. The accuracy is the mean value of applying ACON and SASA as the baselines.
| Cost | Cost Hungarian | IDSA Accuracy | Original |
|---|---|---|---|
| logpsd | 1.00 | 89.6 | 46.9 |
| pshape | 0.50 | 46.4 | 46.9 |
| combo | 0.56 | 52.2 | 46.9 |
| eq6 | 0.25 | 56.0 | 46.9 |
The results suggest that IDSA's effectiveness depends on the quality of the cost function. In particular, logpsd achieves a Hungarian matching cost of 1 and an accuracy of 89.6%, whereas alternative costs with higher matching costs yield smaller improvements over the Original performance. These findings indicate that IDSA accommodates different cost functions, provided they capture meaningful sensor-wise relationships. Accordingly, we will clarify that Eq. (6) is one possible choice for constructing the sensor-wise cost matrix, rather than a requirement, and that alternative costs based on other sensor characteristics can also be incorporated.
Heterogeneous measurement and phase variation
We further investigated the robustness of IDSA under heterogeneous measurement and phase variation. For heterogeneous measurement, we introduced an additional constant offset to the signals of a specific target domain, considering offset magnitudes of 1, 10, and 100. For phase variation, we shifted the temporal signals in the target domain. We evaluated each scenario both with and without sensor permutation to examine the combined effects of measurement variations and changes in sensor correspondence.
ACON
| Permute / model | No / orig | No / IDSA | Yes / orig | Yes / IDSA | Hungarian Cost |
|---|---|---|---|---|---|
| Offset (100) | 5.3 | 7.2 | 6.5 | 6.2 | 1.0 |
| Offset (10) | 27.3 | 29.0 | 11.0 | 28.0 | 1.0 |
| Offset (1) | 71.8 | 72.0 | 46.1 | 64.2 | 1.0 |
| Shift | 88.0 | 88.6 | 57.4 | 87.5 | 1.0 |
SASA
| Permute / model | No / orig | No / IDSA | Yes / orig | Yes / IDSA | Hungarian Cost |
|---|---|---|---|---|---|
| Offset (100) | 6.0 | 6.9 | 5.5 | 5.7 | 1.0 |
| Offset (10) | 25.7 | 25.5 | 13.7 | 17.3 | 1.0 |
| Offset (1) | 73.3 | 77.8 | 33.5 | 73.6 | 1.0 |
| Shift | 91.3 | 92.1 | 36.7 | 91.5 | 1.0 |
IDSA has a limited effect under drastic measurement offsets (e.g., 100), where distribution discrepancies are substantial, but provides notable improvements under moderate offsets combined with sensor permutation. For example, at offsets of 10 and 1, ACON improves from 11.0% to 28.0% and from 46.1% to 64.2%, respectively, while SASA improves from 13.7% to 17.3% and from 33.5% to 73.6%.
Under phase variation combined with sensor permutation, IDSA also substantially recovers performance. Overall, these results suggest that IDSA can mitigate sensor-wise discrepancies arising from changes in sensor correspondence, including those accompanied by moderate measurement offsets or temporal phase variations, although its effectiveness is limited under severe measurement offsets.
Activity Proportions
To investigate the effect of source–target class-distribution differences, we conduct a controlled experiment on SD-gesture by sampling 100 instances from each domain to construct synthetic binary classification settings. We vary the source class proportions as r:(1-r) and reverse them in the target domain to (1-r):r, thereby examining IDSA under different degrees of class-distribution discrepancy. The resulting accuracy (%) is reported below.
| SD-gesture | r | Original – no IDSA | IDSA |
|---|---|---|---|
| ACON | 0.5 | 86.0 | 88.2 |
| ACON | 0.4 | 78.4 | 79.6 |
| ACON | 0.3 | 60.6 | 61.4 |
| ACON | 0.2 | 31.4 | 29.6 |
| SASA | 0.5 | 96.6 | 98.6 |
| SASA | 0.4 | 93.8 | 95.2 |
| SASA | 0.3 | 79.2 | 77.4 |
| SASA | 0.2 | 61.2 | 57.8 |
IDSA achieves comparable performance under moderate distribution differences (r=0.4), but slightly underperforms the baseline under larger differences (r=0.3 for SASA and r=0.2 for ACON). This may be because the mini-batch optimization of Eq. 6 uses averaged representations to estimate sensor-wise dependencies, which can be affected by differing class compositions across source and target batches.
We note that this finding arises from a synthetic setting with explicitly reversed class proportions. In contrast, WISDM exhibits natural class imbalance across subjects (Appendix A.7.1), yet IDSA improves performance in the main evaluation, suggesting that mitigating sensor-wise domain shifts remains beneficial under realistic class-imbalance conditions.
Response to Critical Weakness 3 (Incomplete mechanistic validation).
We thank the reviewer for this helpful comment. We acknowledge that the original description of the PCA distributions did not provide sufficient details, and the experimental design in Appendix A.5.2 incorrectly described the transformed target signals as weighted combinations of source signals. In the actual diagnostic procedure, source and target sensor signals are used to estimate cross-domain relations, while the weighted aggregation is performed exclusively over target-domain signals. In the revised manuscript, we have corrected and expanded the explanation to clarify how the cross-domain relations and aggregation are constructed. Specifically, source-domain signals are used to determine the cross-domain relations, whereas the signals being aggregated are exclusively drawn from the target domain. We further clarify that this motivating example illustrates a naive approach that follows a similar underlying principle to our method, in which cross-domain relations are established using source-domain information and subsequently applied to target-domain signals. We believe these additional explanations provide a clearer understanding of both the experimental setup and its connection to our proposed approach.
Additionally, to further examine the effectiveness of our method design, we conduct additional experiments comparing the learnable inter-domain sensor relation matrix against two alternative sensor-matching strategies: (1) identity mapping, which replaces with an identity matrix , assuming direct one-to-one sensor correspondence; and (2) Hungarian assignment, which establishes one-to-one sensor correspondence through dataset-level assignment.
1. Identity mapping ().
Nina18
| Model | Original | Change Identity |
|---|---|---|
| IDSA + Deep Coral | 85.53 | 81.25 |
| IDSA + CLUDA | 81.88 | 75.76 |
| IDSA + ACON | 83.77 | 83.57 |
| IDSA + SASA | 84.32 | 84.22 |
Opportunity HAR
| Model | Original | Change Identity |
|---|---|---|
| IDSA + Deep Coral | 81.43 | 75.63 |
| IDSA + CLUDA | 84.68 | 80.25 |
| IDSA + ACON | 81.96 | 73.99 |
| IDSA + SASA | 82.96 | 71.16 |
The results show that replacing the learned with an identity matrix generally reduces performance across both Nina5-18 and Opportunity HAR. The performance degradation is particularly noticeable on Opportunity HAR, where accuracy decreases by 4.43–11.80 percentage points across the evaluated backbones. These results suggest that assuming direct sensor correspondence may be insufficient to capture the sensor-wise variations between domains.
2. Hungarian assignment.
Nina18
| Model | Original | Change Assignment |
|---|---|---|
| IDSA + Deep Coral | 85.53 | 75.70 |
| IDSA + CLUDA | 81.88 | 81.80 |
| IDSA + ACON | 83.77 | 82.88 |
| IDSA + SASA | 84.32 | 82.36 |
Opportunity HAR
| Model | Original | Change Assignment |
|---|---|---|
| IDSA + Deep Coral | 81.43 | 69.62 |
| IDSA + CLUDA | 84.68 | 74.70 |
| IDSA + ACON | 81.96 | 79.38 |
| IDSA + SASA | 82.96 | 76.85 |
We further compare IDSA with Hungarian assignment, which provides an alternative based on explicit one-to-one direct sensor matching. The results show that the learned achieves higher accuracy across all reported comparisons on both datasets. In particular, the performance differences are more pronounced on Opportunity HAR, where Hungarian assignment leads to substantial degradation for both DeepCORAL and CLUDA.
3. Simple Linear Adapter.
| Orig | IDSA | IDSA-MLP | |
|---|---|---|---|
| Nina18-deepcoral | 83.47 | 85.53 | 79.47 |
| Nina18-acon | 82.07 | 83.77 | 82.05 |
| Nina18-sasa | 83.06 | 84.32 | 82.41 |
| Nina18-cluda | 80.64 | 81.88 | 79.97 |
| Oppo-deepcoral | 78.27 | 81.43 | 75.50 |
| Oppo-acon | 77.59 | 81.96 | 79.33 |
| Oppo-sasa | 80.51 | 82.96 | 80.69 |
| Oppo-cluda | 82.20 | 84.68 | 84.44 |
| SD-deepcoral | 73.02 | 80.15 | 61.22 |
| SD-acon | 76.96 | 80.80 | 77.30 |
| SD-sasa | 75.45 | 77.01 | 74.82 |
| SD-cluda | 72.49 | 75.68 | 53.68 |
| WISDM-deepcoral | 62.01 | 75.63 | 73.42 |
| WISDM-acon | 77.61 | 79.89 | 79.37 |
| WISDM-sasa | 73.94 | 77.51 | 76.11 |
| WISDM-cluda | 75.98 | 82.14 | 74.0 |
From the table above, replacing the with an MLP layer consistently degrades performance, suggesting that the proposed IDSA design is effective in mitigating sensor-wise domain shifts.
Taken together, these experiments suggest that learning provides advantages over predefined identity correspondence and dataset-level one-to-one sensor assignment. Unlike these alternatives, IDSA learns inter-domain sensor relationships that are not restricted to fixed or one-to-one correspondence. This flexibility appears beneficial for capturing sensor-wise variations across domains and supports the design choice of using a learnable sensor relation matrix in IDSA.
We have incorporated these additional experiments into the revised manuscript to provide further evidence of the contribution of the proposed inter-domain sensor relation learning mechanism.
Response to Critical Weakness 1 (Restricted alignment capacity).
We thank the reviewer for the constructive feedback. We address each point below.
We acknowledge that some of our original statements regarding the transformations that IDSA can handle were broad. To clarify the scope of our method, we conduct additional controlled experiments using synthetically generated datasets, where specific sensor-wise transformations can be independently introduced.
Experimental Results
As discussed in the motivation section, IDSA is primarily designed to address sensor-wise variations involving channel permutation and changes in sensor correspondence. The results support this design objective: under channel permutation (P), IDSA substantially improves accuracy from 37.5% to 92.9%, approaching the performance of the aligned setting (9%). Furthermore, the results under the dead-channel setting (D) suggest that IDSA can maintain or recover useful predictive information even when some sensor channels become unavailable.
However, the results also reveal an important limitation. Under sign-flip transformations (F and PF), where the target sensor signals are inverted (i.e., −x instead of x), IDSA does not provide the expected improvements and performs worse than the original models. We hypothesize that this limitation is related to the formulation of the inter-domain sensor transformation, which represents target sensor information through combinations of existing sensor signals based on source-target relations. Although this formulation can capture changes in sensor correspondence, it may not adequately accommodate transformations involving sign inversion.
These findings help distinguish the transformations that IDSA can effectively address from those that remain challenging for its current formulation. Specifically, IDSA demonstrates strong performance under channel permutation and promising behavior under dead-channel conditions, whereas sign-flip transformations remain a limitation. We acknowledge that our original description did not sufficiently distinguish these cases and will revise the manuscript to clarify the scope and assumptions of IDSA accordingly.
Additionally, to directly address the reviewer’s request, we evaluate Deep CORAL with IDSA under the three transformation settings used in the original diagnostic analysis in Fig. 2(a), namely sensor orientation, sensor placement, and channel permutation. The results show improvements over Deep CORAL in most permutation settings and across several sensor placement conditions. Under axis rotations, however, the gains are mixed. IDSA improves upon Deep CORAL at 108°, 144°, 180°, and 216°, but reduces accuracy at the two smallest tested rotation angles 36° and 72°. Its accuracy also varies less across the tested rotation angles than that of either SASA or Deep CORAL. These results demonstrate that the practical benefit of IDSA depends on the transformation family and that its effectiveness under channel permutation does not imply consistent improvements under general orientation changes. We have updated Fig. 2(a) in the main manuscript to include these additional results and revised the accompanying discussion in Section 4.2 accordingly.