Pith. sign in

REVIEW 4 major objections 6 minor 62 references

Learning Latent Spaces for Domain Generalization in Time Series Forecasting

T0 review · 4 major / 6 minor · reviewed 2026-08-11 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read A plug-in framework called LTG learns latent temporal factors and reports better forecasting accuracy on unseen domains than state-of-the-art domain-generalization methods.

desk verdict Reasonable plug-in for time series DG with mixed evidence; the decoder-conditioning mechanism needs direct verification before the 'overall best' claim can hold. read the letter →

arxiv 2412.11171 v1 pith:PLKMRSBG submitted 2024-12-15 cs.LG

Songgaojun Deng , Maarten de Rijke This is my paper · ORCID
classification cs.LG
keywords timeseriesforecastingdomaingeneralizationvariationalautoencoderlatentspacedisentanglementconditionalVAEshiftprobabilistic
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

Time series forecasting models often fail when asked to predict on new but related domains, such as a website or store never seen during training. This paper argues that existing domain-generalization methods align feature distributions or use label information but ignore the latent factors that actually govern temporal dependencies. It introduces LTG, a plug-in framework that learns these latent factors with a conditional β-variational autoencoder on trend and seasonal components, splitting the latent vector into shared and domain-specific parts. The paper reports that LTG improves range accuracy on unseen domains over strong baselines across five real-world datasets, and that it can be attached to different base forecasters with modest added training cost.

What carries the argument

The load-bearing object is the Conditional $\beta$-VAE: an autoencoder whose encoder estimates a latent posterior $q_\phi(z|x)$ for each decomposed component, and whose decoder reconstructs the component from $z$ together with a one-hot domain identifier $\mathrm{DomID}$. Because the domain ID enters only the decoder, the encoder stays domain-agnostic and the latent vector must carry both shared and domain-specific information; a domain-regularization loss then splits $z$ into a shared part and a specific part by forcing shared parts of all samples close together and specific parts of different domains apart. A two-stage procedure pretrains the VAE and then fine-tunes its encoder while training any base forecaster on the augmented input $x' = \mathrm{Concat}(z,x)W + b$.

What would settle it

Train LTG, then at test time replace every latent vector with a version from which the supposed domain-specific half has been removed (or shuffled across domains); if quantile loss on unseen domains does not change, the domain-specific mechanism carries no information and the claimed separation is not occurring. A companion check: permute the DomID fed to the decoder during pretraining and measure whether downstream forecasting accuracy drops.

Watch

Extended reading notes

Core claim

On the paper's own terms, the central discovery is that modeling temporal dependencies in a disentangled latent space—rather than only matching feature distributions across domains—is what lets a forecaster generalize to unseen domains. LTG's conditional β-VAE learns latent vectors whose shared part encodes domain-invariant patterns and whose specific part encodes per-domain behavior, and feeding both to a forecasting decoder improves range accuracy (quantile loss) in most settings. The paper argues that prior approaches neglect these latent temporal dependencies, and demonstrates the point by showing improved performance across web traffic, retail sales, stock volume, and power consumption data.

Load-bearing premise

The weakest load-bearing premise is that conditioning the decoder on the domain ID during pretraining actually injects useful domain-specific information into the latent vector, even though the ID is absent at test time and only the encoder is used; if the domain signal lives entirely in the decoder, the forecasting gains would have to come from the shared latent part alone.

Editorial extensions

If this is right

  • LTG can be bolted onto DeepAR, WaveNet, DLinear, and GPT4TS, and in most configurations it lowers quantile loss on unseen domains compared with the base model alone.
  • The domain-regularized latent space separates domains visually: t-SNE plots show overlapping shared parts and separated specific parts across test domains.
  • Two-stage training is necessary; training the whole pipeline end-to-end worsens results, so the latent factors must be established before the forecaster uses them.
  • Using only shared latent vectors can outperform the full model on some datasets, indicating that domain-specific factors are not always beneficial.
  • LTG's pretraining is cheaper than the gradient-matching baseline IDGM, and total runtime stays close to the base model.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • If the domain signal does persist in the latent vector despite being absent at test time, LTG is effectively learning a domain code; a direct comparison against simply concatenating the one-hot domain ID to the forecaster input would isolate how much of the gain is from the VAE machinery versus the ID itself.
  • The hyperparameters β and the split fraction α jointly control how much information the specific half can hold; an automatic selection of α per dataset could make LTG more robust, since the Web-traffic result suggests the optimal split varies.
  • Because latent factors are only used as extra input, LTG's gains are bounded by the base model's capacity to use them; the paper's own stated limitation opens the door to using the latent factors to condition the forecaster's parameters or training objective instead.
  • The method is univariate only; extending to multivariate series would require deciding whether to share latent factors across channels, which is a natural next test.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 6 minor

Summary. This paper proposes Latent Temporal Generalization (LTG), a plug-in framework for domain generalization in probabilistic time series forecasting. LTG first decomposes each input series into trend-cyclical and seasonal components, models each component with a conditional beta-VAE whose decoder is conditioned on a domain identifier, and adds a domain-regularization term that encourages part of the latent vector to be shared across domains and the remainder to be domain-specific. The resulting latent vector is concatenated with the raw input through a linear layer and fed into an arbitrary forecasting decoder, which is trained in a second stage while the VAE encoders are fine-tuned. Experiments on five datasets (web traffic, retail, finance, energy) compare LTG applied to DeepAR, WaveNet, DLinear, and GPT4TS against IDGM, Cedar, and UniTime. The paper reports improved range accuracy in several configurations, especially with WaveNet, and includes ablations, t-SNE latent-space visualizations, a forecast-horizon sensitivity study, and a runtime analysis.

Significance. If the main empirical claim were fully established, LTG would be a useful and modular contribution: it is architecture-agnostic, requires no label information, and adds only a lightweight pretraining stage. The experimental design is broad, covering four base forecasters and five real-world datasets, and the two-stage training scheme is simple to implement. However, the claim that LTG 'achieves the overall best performance' is currently supported only by selected comparisons of means over five seeds; several configurations show performance degradation, and no significance tests or aggregate rankings are reported. The central mechanism, namely that domain-specific information is preserved in the latent vector after the DomID-conditioned decoder is removed, is not verified, and the latent-space visualization confirms an objective that is explicitly enforced by Equation (9), so it does not independently validate disentanglement. These gaps make the evidence for the core claim conditional rather than conclusive.

major comments (4)
  1. [Section 5.2, Tables 2 and 3] The paper's headline claim that LTG achieves the overall best range accuracy is not substantiated by the reported results. On Web-traffic, LTG degrades both base models: DeepAR Q(0.5) increases from 0.155 +/- 0.030 to 0.172 +/- 0.014 and Q(mean) from 0.122 +/- 0.018 to 0.179 +/- 0.038; DLinear Q(0.5) increases from 0.152 +/- 0.003 to 0.172 +/- 0.032 and Q(mean) from 0.119 +/- 0.008 to 0.145 +/- 0.040. The Power-cons dataset, where LTG shows its largest gains, has only three domains total, so the 20% test split leaves one test domain per random split. No significance tests, confidence intervals, or win-rate/rank summaries are reported, so it is unclear whether the differences over five seeds support the conclusion. Please add paired significance tests across seeds, report per-configuration wins and losses, and discuss the small number of test domains.
  2. [Section 4.2 and Table 4] The proposed decoder-conditioning mechanism is not verified at inference time. The text states that 'we only need the encoder to generate the latent vectors' at test time, and the encoder is not given DomID; in a VAE the decoder conditions z through reconstruction, so the model can satisfy its training objective by moving domain-specific information into the decoder and leaving z largely shared. The domain-regularization loss in Eq. (9) softly encourages zspecific to differ across domains, but no quantitative measurement (e.g., a domain classifier trained on zspecific, or mutual information between zspecific and DomID) demonstrates that the encoder-only latent vectors actually carry domain-specific information. The ablation LTG(w/o Cond) is inconsistent with the proposed mechanism: removing DomID from the decoder improves DeepAR on Web-traffic Q(0.5) from 0.1717 to 0.1635, while degrading some Stock-volume metrics. The authors' own limitation statement in Section 6 that latent vectors are used only as additional input knowledge further suggests the mechanism is not load-bearing. Please provide a direct test of the information content of zspecific as used at inference.
  3. [Section 5.4.1 and Equation (9)] The t-SNE visualizations are presented as evidence that domain-shared and domain-specific components are separated, but Equation (9) explicitly imposes this separation: it minimizes pairwise distances among zshared samples and maximizes pairwise distances between zspecific samples from different domains. The figure therefore confirms the regularizer rather than providing independent evidence. The disentanglement claim in the abstract and contributions should be supported with a quantitative measure (e.g., MIG or DCI on the learned latents, or an evaluation of whether zspecific predicts domain labels and zshared does not).
  4. [Equations (6)-(8)] The pretraining objective as written is ambiguous. Equation (6) includes an L2 reconstruction term for (xhat_t + xhat_s) - x, and Equations (7)-(8) add log p_theta_t(xt|zt,DomID) and log p_theta_s(xs|zs,DomID) as separate quantities. If the log p terms are Gaussian log-likelihoods, they are themselves reconstruction terms up to scaling, so the objective double-counts the same fit; if they are intended to be the ELBO reconstruction terms, then Equation (6) duplicates them. This makes the exact training signal in Stage 1 (Section 4.5) unclear and is an obstacle to reproducing the method. Please rewrite the objective to specify one reconstruction term and state how the L2 term and the log-likelihood terms are combined.
minor comments (6)
  1. [Table 2] Several entries in Table 2 use a comma instead of a period as the decimal separator (e.g., '0.016, .001'), and the spacing of the mean and standard deviation is inconsistent across columns; please unify the formatting.
  2. [Section 4.3] The definition of the split index as index = alpha * dz should specify how non-integer values are handled (e.g., floor or round), since alpha values such as 0.25 with odd dimension dz are used in the hyperparameter search.
  3. [Appendix E.4, Table 5] The TimeLLM rows in Table 5 are missing Q(0.5) values for Favorita-cat and Stock-volume; please explain why those entries are absent or provide the values.
  4. [Section 4.5 and Abstract] There are typographical issues, including 'Contidional' in the Section 4.5 heading and inconsistent spacing in 'V AE' and 'beta-V AE' in the abstract and introduction.
  5. [Appendix C] The phrase 'reverse instance norm' should be 'reversible instance normalization', and the description should state whether the normalization is applied to the input of the VAE, the forecasting decoder, or both.
  6. [General] No code is released, and the paper does not mention a plan to do so; given the number of hyperparameters and the two-stage training procedure, releasing code or a detailed pseudocode would substantially improve reproducibility.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: held-out test-domain evaluation grounds the central claim; self-citations to [11] for datasets, baselines, and protocols are not load-bearing.

full rationale

The paper's central claim, that LTG improves domain generalization in time series forecasting, is evaluated on held-out test domains in Tables 2 and 3, so the main result is externally grounded rather than derived from its own assumptions. The method's components—decomposition, Conditional β-VAE, domain regularization, and fine-tuning—are tested through ablations on held-out domains, and the best configuration is selected via training-domain validation. The latent-space t-SNE analysis in Section 5.4.1 is the closest to a circular check, since Eq. 9 directly encourages shared-latent similarity and specific-latent dissimilarity; however, the visualization is performed on held-out test domains and is presented as qualitative analysis, not as the basis for the forecasting performance claim. Self-citations to [11] for the common-patterns assumption, dataset preprocessing, the Cedar baseline, and model-selection protocol are pragmatic references to the authors' prior work rather than a uniqueness theorem or an unverified load-bearing premise; moreover, the paper compares against Cedar on held-out domains, giving the comparison independent content. No fitted parameter is renamed as a prediction, and no load-bearing result reduces by construction to its own inputs.

Assumptions & free parameters 5 free parameters · 4 assumptions · 0 invented entities

The central method rests on several searched hyperparameters and domain assumptions. The learned latent factors are not new physical entities but model constructs. The most consequential assumption is that shared temporal patterns transfer across domains, which is standard for DG but not guaranteed on the test splits used.

free parameters (5)
  • beta (VAE capacity weight) = searched from {1, 5, 10, 15}
    Controls KL weight in Eqs. 7-8; chosen by validation performance, not derived.
  • alpha (shared/specific split ratio) = searched from {0.25, 0.5, 0.75}
    Sets index = alpha * dz in Eq. 9; selection changes what is considered shared versus specific.
  • moving average kernel size = searched from {5, 9, 13}
    Defines the trend decomposition in Eq. 1; tuned per dataset.
  • latent dimension dz = inherited from base model hidden size
    Not independently tuned; depends on the forecasting backbone.
  • learning rate and hidden size = learning rate {0.0001, 0.0005, 0.001}; hidden {16, 32, 64}
    Standard hyperparameters searched as described in Appendix C.
assumptions (4)
  • domain assumption Common underlying patterns assumption: shared patterns across source domains transfer to unseen domains
    Explicitly stated in Section 3 and borrowed from [11]; if violated, no domain generalization method can succeed.
  • standard math Standard VAE reparameterization and Gaussian prior for latents
    Used in Eqs. 3-4; standard background for variational inference.
  • domain assumption Training-domain validation is a reliable model selection criterion for domain generalization
    Appendix C cites Gulrajani and Lopez-Paz [20]; this is an assumption adopted without verification on these datasets.
  • ad hoc to paper A fixed split ratio alpha cleanly separates shared from specific latent factors
    Eq. 9 imposes a hard index boundary; there is no evidence that a fixed ratio is sufficient across datasets.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Learning Latent Spaces for Domain Generalization in Time Series Forecasting." pith.science (2026). https://pith.science/paper/PLKMRSBG

@misc{pith2026241211171,
  author       = {Pith},
  title        = {Pith review of: Learning Latent Spaces for Domain Generalization in Time Series Forecasting},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/PLKMRSBG}},
  note         = {Machine review of arXiv:2412.11171}
}
abstract

Time series forecasting is vital in many real-world applications, yet developing models that generalize well on unseen relevant domains -- such as forecasting web traffic data on new platforms/websites or estimating e-commerce demand in new regions -- remains underexplored. Existing forecasting models often struggle with domain shifts in time series data, as the temporal patterns involve complex components like trends, seasonality, etc. While some prior work addresses this by matching feature distributions across domains or disentangling domain-shared features using label information, they fail to reveal insights into the latent temporal dependencies, which are critical for identifying common patterns across domains and achieving generalization. We propose a framework for domain generalization in time series forecasting by mining the latent factors that govern temporal dependencies across domains. Our approach uses a decomposition-based architecture with a new Conditional $\beta$-Variational Autoencoder (VAE), wherein time series data is first decomposed into trend-cyclical and seasonal components, each modeled independently through separate $\beta$-VAE modules. The $\beta$-VAE aims to capture disentangled latent factors that control temporal dependencies across domains. We enhance the learning of domain-specific information with a decoder-conditional design and introduce domain regularization to improve the separation of domain-shared and domain-specific latent factors. Our proposed method is flexible and can be applied to various time series forecasting models, enabling effective domain generalization with simplicity and efficiency. We validate its effectiveness on five real-world time series datasets, covering web traffic, e-commerce, finance and power consumption, demonstrating improved generalization performance over state-of-the-art methods.

Figures

Figures reproduced from arXiv: 2412.11171 by the authors.

Figure 1
Figure 1. The proposed framework, LTG, consists of three components: (1) Time Series Decomposition, which decomposes a raw time sequence into trend-cyclical and seasonal components for more effective modeling of distinct temporal patterns. (2) Latent Factor Learning with Conditional β-VAE, which learns latent representations that capture temporal dependencies across different domains and ideally achieves disentangled latent f… view at source ↗
Figure 2
Figure 2. Forecasting results for a test domain sample from the Web-traffic dataset, using DLinear and WaveNet with [PITH_FULL_IMAGE:figures/full_fig_p010_2.png] view at source ↗
Figure 3
Figure 3. Latent space t-SNE visualization of the domain-shared and domain-specific components on the test domains [PITH_FULL_IMAGE:figures/full_fig_p011_3.png] view at source ↗
Figures from the paper (5 more)
Figure 4
Figure 4. Figure 4: Forecasting window sensitivity and runtime analysis on the Web-traffic dataset using DeepAR as the base [PITH_FULL_IMAGE:figures/full_fig_p012_4.png]
Figure 5
Figure 5. Figure 5: Forecasting results of Q(0.5) and Q(mean) on the Web-traffic dataset with varying values of [PITH_FULL_IMAGE:figures/full_fig_p016_5.png]
Figure 6
Figure 6. Figure 6: Latent space t-SNE visualization of the domain-shared and domain-specific components on the test domains [PITH_FULL_IMAGE:figures/full_fig_p017_6.png]
Figure 7
Figure 7. Figure 7: Forecasting results for test domain samples from the Web-traffic dataset, using DLinear and WaveNet with [PITH_FULL_IMAGE:figures/full_fig_p017_7.png]
Figure 8
Figure 8. Figure 8: Forecasting results for a test domain sample from the Stock-volume dataset, using DeepAR with various [PITH_FULL_IMAGE:figures/full_fig_p018_8.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

62 extracted references · 42 canonical work pages

  1. [1]

    A. F. Ansari, L. Stella, C. Turkmen, X. Zhang, P. Mercado, H. Shen, O. Shchur, S. S. Rangapuram, S. P. Arango, S. Kapoor, et al. Chronos: Learning the language of time series. arXiv preprint arXiv:2403.07815, 2024

    arXiv 2024

  2. [2]

    J. S. Armstrong. Long-Range Forecasting: From Crystal Ball to Computer. Wiley, 1978

    work page 1978

  3. [3]

    Bahdanau, K

    D. Bahdanau, K. Cho, and Y . Bengio. Neural machine translation by jointly learning to align and translate. arXiv preprint arXiv:1409.0473, 2014

    arXiv 2014

  4. [4]

    G. Bai, C. Ling, and L. Zhao. Temporal domain generalization with drift-aware dynamic neural networks. In The Eleventh International Conference on Learning Representations, 2023. URL https://openreview.net/ forum?id=sWOsRj4nT1n

    work page 2023

  5. [5]

    S. Bai, J. Z. Kolter, and V . Koltun. An empirical evaluation of generic convolutional and recurrent networks for sequence modeling. arXiv preprint arXiv:1803.01271, 2018

    arXiv 2018

  6. [6]

    Balaji, S

    Y . Balaji, S. Sankaranarayanan, and R. Chellappa. MetaReg: Towards domain generalization using meta- regularization. Advances in Neural Information Processing Systems, 31, 2018

    work page 2018

  7. [7]

    M. A. Benjamin, R. A. Rigby, and D. M. Stasinopoulos. Generalized autoregressive moving average models. Journal of the American Statistical Association, 98(461):214–223, 2003

    work page 2003

  8. [8]

    C. P. Burgess, I. Higgins, A. Pal, L. Matthey, N. Watters, G. Desjardins, and A. Lerchner. Understanding disentangling in beta-V AE.arXiv preprint arXiv:1804.03599, 2018

    arXiv 2018

Show all 62 references
  1. [9]

    Chung, C

    J. Chung, C. Gulcehre, K. Cho, and Y . Bengio. Empirical evaluation of gated recurrent neural networks on sequence modeling. arXiv preprint arXiv:1412.3555, 2014

    2014 arXiv

  2. [10]

    R. B. Cleveland, W. S. Cleveland, J. E. McRae, and I. Terpenning. STL: A seasonal-trend decomposition. J. of Official Statistics, 6(1):3–73, 1990

    1990

  3. [11]

    S. Deng, O. Sprangers, M. Li, S. Schelter, and M. de Rijke. Domain generalization in time series forecasting. ACM Transactions on Knowledge Discovery from Data, 18(5):1–24, 2024

    2024

  4. [12]

    Y . Du, J. Wang, W. Feng, S. Pan, T. Qin, R. Xu, and C. Wang. AdaRNN: Adaptive learning and forecasting of time series. In Proceedings of the 30th ACM international Conference on Information & Knowledge Management, pages 402–411, 2021

    2021

  5. [13]

    W. Fan, P. Wang, D. Wang, D. Wang, Y . Zhou, and Y . Fu. Dish-ts: A general paradigm for alleviating distribution shift in time series forecasting. In Proceedings of the AAAI conference on artificial intelligence, volume 37, pages 7522–7529, 2023

    2023

  6. [14]

    Gagnon-Audet, K

    J.-C. Gagnon-Audet, K. Ahuja, M.-J. Darvishi-Bayazi, P. Mousavi, G. Dumas, and I. Rish. Woods: Benchmarks for out-of-distribution generalization in time series. arXiv preprint arXiv:2203.09978, 2022

    2022 arXiv

  7. [15]

    Ganin, E

    Y . Ganin, E. Ustinova, H. Ajakan, P. Germain, H. Larochelle, F. Laviolette, M. Marchand, and V . Lempitsky. Domain-adversarial training of neural networks. The Journal of Machine Learning Research, 17(1):2096–2030, 2016. 12

    2016

  8. [16]

    E. S. Gardner Jr. Exponential smoothing: The state of the art—part ii. International Journal of Forecasting, 22(4): 637–666, 2006

    2006

  9. [17]

    Glorot and Y

    X. Glorot and Y . Bengio. Understanding the difficulty of training deep feedforward neural networks. InAISTATS, 2010

    2010

  10. [18]

    Gneiting and M

    T. Gneiting and M. Katzfuss. Probabilistic forecasting. Annual Review of Statistics and Its Application, 1:125–151, 2014

    2014

  11. [19]

    R. Gong, W. Li, Y . Chen, and L. V . Gool. DLOW: Domain flow for adaptation and generalization. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 2477–2486, 2019

    2019

  12. [20]

    Gulrajani and D

    I. Gulrajani and D. Lopez-Paz. In search of lost domain generalization. arXiv preprint arXiv:2007.01434, 2020

    2007 arXiv

  13. [21]

    J. D. Hamilton. Time Series Analysis. Princeton University Press, 2020

    2020

  14. [22]

    Higgins, L

    I. Higgins, L. Matthey, A. Pal, C. P. Burgess, X. Glorot, M. M. Botvinick, S. Mohamed, and A. Lerchner. beta-vae: Learning basic visual concepts with a constrained variational framework. ICLR, 3, 2017

    2017

  15. [23]

    Hochreiter and J

    S. Hochreiter and J. Schmidhuber. Long short-term memory. Neural Computation, 9(8):1735–1780, 1997

    1997

  16. [24]

    Y . Hu, X. Jia, M. Tomizuka, and W. Zhan. Causal-based time series domain generalization for vehicle intention prediction. In 2022 International Conference on Robotics and Automation (ICRA), pages 7806–7813. IEEE, 2022

    2022

  17. [25]

    Hupkes, M

    D. Hupkes, M. Giulianelli, V . Dankers, M. Artetxe, Y . Elazar, T. Pimentel, C. Christodoulopoulos, K. Lasri, N. Saphra, A. Sinclair, et al. A taxonomy and review of generalization research in NLP. Nature Machine Intelligence, 5(10):1161–1174, 2023

    2023

  18. [26]

    M. Ilse, J. M. Tomczak, C. Louizos, and M. Welling. Diva: Domain invariant variational autoencoders. InMedical Imaging with Deep Learning, pages 322–348. PMLR, 2020

    2020

  19. [27]

    M. Jin, S. Wang, L. Ma, Z. Chu, J. Y . Zhang, X. Shi, P.-Y . Chen, Y . Liang, Y .-F. Li, S. Pan, and Q. Wen. Time-LLM: Time series forecasting by reprogramming large language models. In The Twelfth International Conference on Learning Representations, 2024

    2024

  20. [28]

    G. Ke, Q. Meng, T. Finley, T. Wang, W. Chen, W. Ma, Q. Ye, and T.-Y . Liu. LightGBM: A highly efficient gradient boosting decision tree. volume 30, 2017

    2017

  21. [29]

    T. Kim, J. Kim, Y . Tae, C. Park, J.-H. Choi, and J. Choo. Reversible instance normalization for accurate time-series forecasting against distribution shift. In International Conference on Learning Representations, 2021

    2021

  22. [30]

    D. P. Kingma and J. Ba. Adam: A method for stochastic optimization. In International Conference on Learning Representations, volume 5, 2015

    2015

  23. [31]

    D. P. Kingma and M. Welling. Auto-encoding variational Bayes. arXiv preprint arXiv:1312.6114, 2013

    2013 arXiv

  24. [32]

    S. Li, X. Jin, Y . Xuan, X. Zhou, W. Chen, Y .-X. Wang, and X. Yan. Enhancing the locality and breaking the memory bottleneck of transformer on time series forecasting. Advances in Neural Information Processing Systems, 32, 2019

    2019

  25. [33]

    Lim and S

    B. Lim and S. Zohren. Time-series forecasting with deep learning: A survey. Philosophical Transactions of the Royal Society A, 379(2194):20200209, 2021

    2021

  26. [34]

    J. Liu, Z. Shen, Y . He, X. Zhang, R. Xu, H. Yu, and P. Cui. Towards out-of-distribution generalization: A survey. arXiv preprint arXiv:2108.13624, 2021

    2021 arXiv

  27. [35]

    X. Liu, J. Hu, Y . Li, S. Diao, Y . Liang, B. Hooi, and R. Zimmermann. Unitime: A language-empowered unified model for domain time series forecasting. In Proceedings of the ACM on Web Conference 2024, pages 4095–4106, 2024

    2024

  28. [36]

    Z. Liu, M. Cheng, Z. Li, Z. Huang, Q. Liu, Y . Xie, and E. Chen. Adaptive normalization for non-stationary time series forecasting: A temporal slice perspective. Advances in Neural Information Processing Systems, 36, 2024

    2024

  29. [37]

    W. Lu, J. Wang, X. Sun, Y . Chen, and X. Xie. Out-of-distribution representation learning for time series classification. In The Eleventh International Conference on Learning Representations , 2023. URL https: //openreview.net/forum?id=gUZWOE42l6Q

    2023

  30. [38]

    Nasery, S

    A. Nasery, S. Thakur, V . Piratla, A. De, and S. Sarawagi. Training for the future: A simple gradient interpolation loss to generalize along time. In A. Beygelzimer, Y . Dauphin, P. Liang, and J. W. Vaughan, editors,Advances in Neural Information Processing Systems, 2021. URL ...

    2021

  31. [39]

    Y . Nie, N. H. Nguyen, P. Sinthong, and J. Kalagnanam. A time series is worth 64 words: Long-term forecasting with transformers. In The Eleventh International Conference on Learning Representations, 2023. URL https: //openreview.net/forum?id=Jbdc0vTOcol. 13

    2023

  32. [40]

    A. v. d. Oord, S. Dieleman, H. Zen, K. Simonyan, O. Vinyals, A. Graves, N. Kalchbrenner, A. Senior, and K. Kavukcuoglu. Wavenet: A generative model for raw audio. arXiv preprint arXiv:1609.03499, 2016

    2016 arXiv

  33. [41]

    Paszke, S

    A. Paszke, S. Gross, F. Massa, A. Lerer, J. Bradbury, G. Chanan, T. Killeen, Z. Lin, N. Gimelshein, L. Antiga, A. Desmaison, A. Köpf, E. Yang, Z. DeVito, M. Raison, A. Tejani, S. Chilamkurthy, B. Steiner, L. Fang, J. Bai, and S. Chintala. Pytorch: An imperative style, high-per...

    2019

  34. [42]

    Radford, J

    A. Radford, J. Wu, R. Child, D. Luan, D. Amodei, I. Sutskever, et al. Language models are unsupervised multitask learners. OpenAI blog, 1(8):9, 2019

    2019

  35. [43]

    Ragab, Z

    M. Ragab, Z. Chen, W. Zhang, E. Eldele, M. Wu, C.-K. Kwoh, and X. Li. Conditional contrastive domain generalization for fault diagnosis. IEEE Transactions on Instrumentation and Measurement, 71:1–12, 2022

    2022

  36. [44]

    Salinas, V

    D. Salinas, V . Flunkert, J. Gasthaus, and T. Januschowski. Deepar: Probabilistic forecasting with autoregressive recurrent networks. International Journal of Forecasting, 36(3):1181–1191, 2020

    2020

  37. [45]

    Y . Shi, J. Seely, P. H. Torr, N. Siddharth, A. Hannun, N. Usunier, and G. Synnaeve. Gradient matching for domain generalization. arXiv preprint arXiv:2104.09937, 2021

    2021 arXiv

  38. [46]

    Sprangers, S

    O. Sprangers, S. Schelter, and M. de Rijke. Parameter-efficient deep probabilistic forecasting. International Journal of Forecasting, 39(1):332–345, 2023

    2023

  39. [47]

    M. Tan, M. A. Merrill, V . Gupta, T. Althoff, and T. Hartvigsen. Are language models actually useful for time series forecasting? arXiv preprint arXiv:2406.16964, 2024

    2024 arXiv

  40. [48]

    Tibshirani

    R. Tibshirani. Regression shrinkage and selection via the lasso. Journal of the Royal Statistical Society: Series B (Methodological), 58(1):267–288, 1996

    1996

  41. [49]

    Van der Maaten and G

    L. Van der Maaten and G. Hinton. Visualizing data using t-sne. Journal of Machine Learning Research, 9(11), 2008

    2008

  42. [50]

    Vaswani, N

    A. Vaswani, N. Shazeer, N. Parmar, J. Uszkoreit, L. Jones, A. N. Gomez, Ł. Kaiser, and I. Polosukhin. Attention is all you need. Advances in Neural Information Processing Systems, 30, 2017

    2017

  43. [51]

    J. Wang, C. Lan, C. Liu, Y . Ouyang, T. Qin, W. Lu, Y . Chen, W. Zeng, and P. S. Yu. Generalizing to unseen domains: A survey on domain generalization, 2022. URL https://arxiv.org/abs/2103.03097

    2022 arXiv

  44. [52]

    Z. Wang, C. Pei, M. Ma, X. Wang, Z. Li, D. Pei, S. Rajmohan, D. Zhang, Q. Lin, H. Zhang, et al. Revisiting V AE for unsupervised time series anomaly detection: A frequency perspective. In Proceedings of the ACM on Web Conference 2024, pages 3096–3105, 2024

    2024

  45. [53]

    G. Woo, C. Liu, A. Kumar, C. Xiong, S. Savarese, and D. Sahoo. Unified training of universal time series forecasting transformers. arXiv preprint arXiv:2402.02592, 2024

    2024 arXiv

  46. [54]

    H. Wu, J. Xu, J. Wang, and M. Long. Autoformer: Decomposition transformers with auto-correlation for long-term series forecasting. Advances in Neural Information Processing Systems, 34:22419–22430, 2021

    2021

  47. [55]

    X. Yue, Y . Zhang, S. Zhao, A. Sangiovanni-Vincentelli, K. Keutzer, and B. Gong. Domain randomization and pyramid consistency: Simulation-to-real generalization without accessing target domain data. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pa...

    2019

  48. [56]

    A. Zeng, M. Chen, L. Zhang, and Q. Xu. Are transformers effective for time series forecasting? In Proceedings of the AAAI conference on artificial intelligence, volume 37, pages 11121–11128, 2023

    2023

  49. [57]

    Zhang, M

    W. Zhang, M. Ragab, and C.-S. Foo. Domain generalization via selective consistency regularization for time series classification. In 2022 26th International Conference on Pattern Recognition (ICPR), pages 2149–2156. IEEE, 2022

    2022

  50. [58]

    W. X. Zhao, K. Zhou, J. Li, T. Tang, X. Wang, Y . Hou, Y . Min, B. Zhang, J. Zhang, Z. Dong, et al. A survey of large language models. arXiv preprint arXiv:2303.18223, 2023

    2023 arXiv

  51. [59]

    Zhao and S

    Y . Zhao and S. Linderman. Revisiting structured variational autoencoders. In International Conference on Machine Learning, pages 42046–42057. PMLR, 2023

    2023

  52. [60]

    H. Zhou, S. Zhang, J. Peng, S. Zhang, J. Li, H. Xiong, and W. Zhang. Informer: Beyond efficient transformer for long sequence time-series forecasting. In Proceedings of the AAAI conference on artificial intelligence, volume 35, pages 11106–11115, 2021

    2021

  53. [61]

    T. Zhou, Z. Ma, Q. Wen, X. Wang, L. Sun, and R. Jin. Fedformer: Frequency enhanced decomposed transformer for long-term series forecasting. In International conference on machine learning, pages 27268–27286. PMLR, 2022

    2022

  54. [62]

    T. Zhou, P. Niu, L. Sun, and R. Jin. One fits all: Power general time series analysis by pretrained lm. Advances in Neural Information Processing Systems, 36:43322–43355, 2023. 14 A Ethical Considerations The research topic of this paper, domain generalization in time series f...

    2023

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.