Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
112 tokens/sec
GPT-4o
8 tokens/sec
Gemini 2.5 Pro Pro
47 tokens/sec
o3 Pro
5 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

MultiWave: Multiresolution Deep Architectures through Wavelet Decomposition for Multivariate Time Series Prediction (2306.10164v1)

Published 16 Jun 2023 in cs.LG, cs.HC, and eess.SP

Abstract: The analysis of multivariate time series data is challenging due to the various frequencies of signal changes that can occur over both short and long terms. Furthermore, standard deep learning models are often unsuitable for such datasets, as signals are typically sampled at different rates. To address these issues, we introduce MultiWave, a novel framework that enhances deep learning time series models by incorporating components that operate at the intrinsic frequencies of signals. MultiWave uses wavelets to decompose each signal into subsignals of varying frequencies and groups them into frequency bands. Each frequency band is handled by a different component of our model. A gating mechanism combines the output of the components to produce sparse models that use only specific signals at specific frequencies. Our experiments demonstrate that MultiWave accurately identifies informative frequency bands and improves the performance of various deep learning models, including LSTM, Transformer, and CNN-based models, for a wide range of applications. It attains top performance in stress and affect detection from wearables. It also increases the AUC of the best-performing model by 5% for in-hospital COVID-19 mortality prediction from patient blood samples and for human activity recognition from accelerometer and gyroscope data. We show that MultiWave consistently identifies critical features and their frequency components, thus providing valuable insights into the applications studied.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (62)
  1. Defining accelerometer thresholds for physical activity in girls using roc analysis. Journal of Physical Activity and Health, 7(1):45–53, 2010.
  2. An autoencoder wavelet based deep neural network with attention mechanism for multi-step prediction of plant growth. Information Sciences, 560:35–50, 2021.
  3. Combining deep learning and multiresolution analysis for stock market forecasting. IEEE Access, 9:13099–13111, 2021.
  4. On multi-rate fusion for non-linear sampled-data systems: Application to a 6d tracking system. Robotics and Autonomous Systems, 56(8):706–715, 2008.
  5. mhealthdroid: a novel framework for agile development of mobile health applications. In Ambient Assisted Living and Daily Activities: 6th International Work-Conference, IWAAL 2014, Belfast, UK, December 2-5, 2014. Proceedings 6, pages 91–98. Springer, 2014.
  6. Design, implementation and validation of a novel open framework for agile development of mobile health applications. Biomedical engineering online, 14(2):1–20, 2015.
  7. Lukas Biewald. Experiment tracking with weights and biases, 2020. URL https://www.wandb.com/. Software available from wandb.com.
  8. The Fourier transform and its applications, volume 31999. McGraw-Hill New York, 1986.
  9. Shimmer™–a wireless sensor platform for noninvasive biomedical research. IEEE Sensors Journal, 10(9):1527–1534, 2010.
  10. Tdstf: Transformer-based diffusion probabilistic model for sparse time series forecasting. arXiv preprint arXiv:2301.06625, 2023.
  11. Recurrent neural networks for multivariate time series with missing values. Scientific reports, 8(1):1–12, 2018a.
  12. Hierarchical deep generative models for multi-rate multivariate time series. In International Conference on Machine Learning, pages 784–793. PMLR, 2018b.
  13. Deep learning for sensor-based human activity recognition: Overview, challenges, and opportunities. ACM Computing Surveys (CSUR), 54(4):1–40, 2021.
  14. Biorthogonal bases of compactly supported wavelets. Communications on pure and applied mathematics, 45(5):485–560, 1992.
  15. Multi-scale convolutional neural networks for time series classification. arXiv preprint arXiv:1603.06995, 2016.
  16. Ingrid Daubechies. Ten lectures on wavelets. SIAM, 1992.
  17. Human activity recognition using inertial, physiological and environmental sensors: A comprehensive survey. IEEE access, 8:210816–210836, 2020.
  18. Adapting auxiliary losses using gradient similarity. arXiv preprint arXiv:1812.02224, 2018.
  19. Can we ditch feature engineering? end-to-end deep learning for affect recognition from physiological sensor data. Sensors, 20(22), 2020. ISSN 1424-8220. 10.3390/s20226535. URL https://www.mdpi.com/1424-8220/20/22/6535.
  20. Grant Foster. Wavelets for period analysis of unevenly sampled time series. The Astronomical Journal, 112:1709–1729, 1996.
  21. Risk factors for severe and critically ill covid-19 patients: a review. Allergy, 76(2):428–455, 2021.
  22. Hyperspectral image classification using cnn-enhanced multi-level haar wavelet features fusion network. IEEE Geoscience and Remote Sensing Letters, 19:1–5, 2022.
  23. Alfred Haar. Zur theorie der orthogonalen funktionensysteme. Georg-August-Universitat, Gottingen., 1909.
  24. Spatial-temporal convolutional transformer network for multivariate time series forecasting. Sensors, 22(3):841, 2022.
  25. Peripheral inflammatory cytokines and lymphocyte subset features of deceased covid-19 patients. BioMed Research International, 2021, 2021.
  26. Forecasting air pollutant concentration using a novel spatiotemporal deep learning model based on clustering, feature selection and empirical wavelet transform. Science of The Total Environment, 801:149654, 2021.
  27. Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980, 2014.
  28. The ‘trier social stress test’–a tool for investigating psychobiological stress responses in a laboratory setting. Neuropsychobiology, 28(1-2):76–81, 1993.
  29. A clockwork rnn. In International Conference on Machine Learning, pages 1863–1871. PMLR, 2014.
  30. Lstm network as a screening tool to detect moderate traumatic brain injury from resting-state electroencephalogram. Expert Systems with Applications, page 116761, 2022.
  31. Fnet: Mixing tokens with fourier transforms. arXiv preprint arXiv:2105.03824, 2021.
  32. A hybrid deep learning framework for long-term traffic flow prediction. IEEE Access, 9:11264–11271, 2021.
  33. Forecasting models for wind speed using wavelet, wavelet packet, time series and artificial neural networks. Applied Energy, 107:191–208, 2013. ISSN 0306-2619. https://doi.org/10.1016/j.apenergy.2013.02.002. URL https://www.sciencedirect.com/science/article/pii/S0306261913001104.
  34. Efficient low-rank multimodal fusion with modality-specific factors. arXiv preprint arXiv:1806.00064, 2018.
  35. Rectified linear units improve restricted boltzmann machines. In Proceedings of the 27th international conference on machine learning (ICML-10), pages 807–814, 2010.
  36. Deep convolutional and lstm recurrent neural networks for multimodal wearable activity recognition. Sensors, 16(1):115, 2016.
  37. Pytorch: An imperative style, high-performance deep learning library. In H. Wallach, H. Larochelle, A. Beygelzimer, F. d'Alché-Buc, E. Fox, and R. Garnett, editors, Advances in Neural Information Processing Systems 32, pages 8024–8035. Curran Associates, Inc., 2019.
  38. Prediction of extubation failure in the paediatric cardiac icu using machine learning and high-frequency physiologic data. Cardiology in the Young, 32(10):1649–1656, 2022.
  39. Multirate multisensor data fusion for linear systems using kalman filters and a neural network. Aerospace Science and Technology, 39:465–471, 2014.
  40. Comparative study of different discrete wavelet based neural network models for long term drought forecasting. Water Resources Management, pages 1–20, 2023.
  41. Introducing wesad, a multimodal dataset for wearable stress and affect detection. In Proceedings of the 20th ACM international conference on multimodal interaction, pages 400–408, 2018.
  42. A survey on principles, models and methods for learning from irregularly sampled time series. arXiv preprint arXiv:2012.00168, 2020.
  43. Forecasting stock market indices using padding-based fourier transform denoising and time series deep learning models. IEEE Access, 9:83786–83796, 2021.
  44. Smart devices are different: Assessing and mitigatingmobile sensing heterogeneities for activity recognition. In Proceedings of the 13th ACM conference on embedded networked sensor systems, pages 127–140, 2015.
  45. Classification of emg signals using wavelet neural network. Journal of Neuroscience Methods, 156(1):360–367, 2006. ISSN 0165-0270. https://doi.org/10.1016/j.jneumeth.2006.03.004. URL https://www.sciencedirect.com/science/article/pii/S0165027006001440.
  46. A review of deep learning methods for irregularly sampled medical time series data. arXiv preprint arXiv:2010.12493, 2020.
  47. Te-esn: Time encoding echo state network for prediction based on irregularly sampled time series data. arXiv preprint arXiv:2105.00412, 2021.
  48. High accuracy at low frequency: detailed behavioural classification from accelerometer data. Journal of Experimental Biology, 221(23), 11 2018. ISSN 0022-0949. 10.1242/jeb.184085. URL https://doi.org/10.1242/jeb.184085. jeb184085.
  49. Self-supervised transformer for multivariate clinical time-series with missing values. arXiv preprint arXiv:2107.14293, 2021.
  50. Deep convolutional neural network with wavelet decomposition for automatic modulation classification. In 2020 15th IEEE Conference on Industrial Electronics and Applications (ICIEA), pages 1566–1571. IEEE, 2020a.
  51. Multilevel wavelet decomposition network for interpretable time series analysis. In Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, pages 2437–2446, 2018.
  52. What makes training multi-modal classification networks hard? In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 12695–12705, 2020b.
  53. Transformers in time series: A survey. arXiv preprint arXiv:2202.07125, 2022.
  54. Machine learning for healthcare: on the verge of a major shift in healthcare epidemiology. Clinical Infectious Diseases, 66(1):149–153, 2018.
  55. A learning algorithm for continually running fully recurrent neural networks. Neural computation, 1(2):270–280, 1989.
  56. An interpretable mortality prediction model for covid-19 patients. Nature machine intelligence, 2(5):283–288, 2020.
  57. A multi-view deep learning method for epileptic seizure detection using short-time fourier transform. In Proceedings of the 8th ACM International Conference on Bioinformatics, Computational Biology, and Health Informatics, pages 213–222, 2017.
  58. Wavelet neural networks for function learning. IEEE transactions on Signal Processing, 43(6):1485–1497, 1995.
  59. Exploiting multi-channels deep convolutional neural networks for multivariate time series classification. Frontiers of Computer Science, 10(1):96–112, 2016.
  60. Fedformer: Frequency enhanced decomposed transformer for long-term series forecasting. In International Conference on Machine Learning, pages 27268–27286. PMLR, 2022.
  61. Waveformer: Linear-time attention with forward and backward wavelet transform. arXiv preprint arXiv:2210.01989, 2022.
  62. An investigation on deep learning and wavelet transform to nowcast wind power and wind power ramp: A case study in brazil and uruguay. Energy, 230:120842, 2021.
Citations (4)

Summary

We haven't generated a summary for this paper yet.