Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
156 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
45 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Continual Learning of Range-Dependent Transmission Loss for Underwater Acoustic using Conditional Convolutional Neural Net (2404.08091v1)

Published 11 Apr 2024 in cs.LG, eess.SP, and physics.flu-dyn

Abstract: There is a significant need for precise and reliable forecasting of the far-field noise emanating from shipping vessels. Conventional full-order models based on the Navier-Stokes equations are unsuitable, and sophisticated model reduction methods may be ineffective for accurately predicting far-field noise in environments with seamounts and significant variations in bathymetry. Recent advances in reduced-order models, particularly those based on convolutional and recurrent neural networks, offer a faster and more accurate alternative. These models use convolutional neural networks to reduce data dimensions effectively. However, current deep-learning models face challenges in predicting wave propagation over long periods and for remote locations, often relying on auto-regressive prediction and lacking far-field bathymetry information. This research aims to improve the accuracy of deep-learning models for predicting underwater radiated noise in far-field scenarios. We propose a novel range-conditional convolutional neural network that incorporates ocean bathymetry data into the input. By integrating this architecture into a continual learning framework, we aim to generalize the model for varying bathymetry worldwide. To demonstrate the effectiveness of our approach, we analyze our model on several test cases and a benchmark scenario involving far-field prediction over Dickin's seamount in the Northeast Pacific. Our proposed architecture effectively captures transmission loss over a range-dependent, varying bathymetry profile. This architecture can be integrated into an adaptive management system for underwater radiated noise, providing real-time end-to-end mapping between near-field ship noise sources and received noise at the marine mammal's location.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (40)
  1. C. M. Duarte, L. Chapuis, S. P. Collin, D. P. Costa, R. P. Devassy, V. M. Eguiluz, C. Erbe, T. A. Gordon, B. S. Halpern, H. R. Harding, et al., “The soundscape of the anthropocene ocean,” Science 371 (2021).
  2. A. Venkateshwaran, I. K. Deo, J. Jelovica,  and R. K. Jaiman, “A multi-objective optimization framework for reducing the impact of ship noise on marine mammals,” arXiv preprint arXiv:2402.02647  (2024).
  3. R. J. LeVeque et al., Finite volume methods for hyperbolic problems, Vol. 31 (Cambridge university press, 2002).
  4. M. B. Porter and H. P. Bucker, “Gaussian beam tracing for computing ocean acoustic fields,” The Journal of the Acoustical Society of America 82, 1349–1359 (1987).
  5. M. B. Porter, “Beam tracing for two-and three-dimensional problems in ocean acoustics,” The Journal of the Acoustical Society of America 146, 2016–2029 (2019).
  6. A. Williams, “Normal-mode methods in propagation of underwater sound,” Underwater acoustics , 23–56 (1970).
  7. D. M. Chapman and P. D. Ward, “The normal-mode theory of air-to-water sound transmission in the ocean,” The Journal of the Acoustical Society of America 87, 601–618 (1990).
  8. K. R. James and D. R. Dowling, “A probability density function method for acoustic field uncertainty analysis,” The Journal of the Acoustical Society of America 118, 2802–2810 (2005).
  9. K. R. James and D. R. Dowling, “Pekeris waveguide comparisons of methods for predicting acoustic field amplitude uncertainty caused by a spatially uniform environmental uncertainty (l),” The Journal of the Acoustical Society of America 129, 589–592 (2011).
  10. Y. LeCun, Y. Bengio,  and G. Hinton, “Deep learning,” nature 521, 436–444 (2015).
  11. I. K. Deo and R. K. Jaiman, “Predicting waves in fluids with deep neural network,” Physics of Fluids  (2022).
  12. I. K. Deo, R. Gao,  and R. Jaiman, “Combined space–time reduced-order model with three-dimensional deep convolution for extrapolating fluid dynamics,” Physics of Fluids 35 (2023).
  13. R. Gao, I. K. Deo,  and R. K. Jaiman, “A finite element-inspired hypergraph neural network: Application to fluid dynamics simulations,” Journal of Computational Physics , 112866 (2024).
  14. C. Yang, X. Yang,  and X. Xiao, “Data-driven projection method in fluid simulation,” Computer Animation and Virtual Worlds 27, 415–424 (2016).
  15. H. Niu, E. Ozanich,  and P. Gerstoft, “Ship localization in santa barbara channel using machine learning classifiers,” The journal of the acoustical society of America 142, EL455–EL460 (2017).
  16. Z. Huang, J. Xu, Z. Gong, H. Wang,  and Y. Yan, “Source localization using deep neural networks in a shallow water environment,” The Journal of the Acoustical Society of America 143, 2922–2932 (2018).
  17. J. Chi, X. Li, H. Wang, D. Gao,  and P. Gerstoft, “Sound source ranging using a feed-forward neural network trained with fitting-based early stopping,” The journal of the acoustical society of America 146, EL258–EL264 (2019).
  18. E. L. Ferguson, “Multitask convolutional neural network for acoustic localization of a transiting broadband source using a hydrophone array,” The Journal of the Acoustical Society of America 150, 248–256 (2021).
  19. S. Fotiadis, E. Pignatelli, M. L. Valencia, C. Cantwell, A. Storkey,  and A. A. Bharath, “Comparing recurrent and convolutional neural networks for predicting wave propagation,” arXiv preprint arXiv:2002.08981  (2020).
  20. N. Borrel-Jensen, A. P. Engsig-Karup,  and C.-H. Jeong, “Physics-informed neural networks for one-dimensional sound field predictions with parameterized sources and impedance boundaries,” JASA Express Letters 1, 122402 (2021).
  21. W. Mallik, R. K. Jaiman,  and J. Jelovica, “Predicting transmission loss in underwater acoustics using convolutional recurrent autoencoder network,” The Journal of the Acoustical Society of America 152, 1627–1638 (2022).
  22. S. R. Bukka, R. Gupta, A. R. Magee,  and R. K. Jaiman, “Assessment of unsteady flow predictions using hybrid deep learning based reduced-order models,” Physics of Fluids 33, 013601 (2021).
  23. L. Wang, X. Zhang, H. Su,  and J. Zhu, “A comprehensive survey of continual learning: Theory, method and application,” IEEE Transactions on Pattern Analysis and Machine Intelligence , 1–20 (2024).
  24. S.-A. Rebuffi, A. Kolesnikov, G. Sperl,  and C. H. Lampert, “icarl: Incremental classifier and representation learning,” in Proceedings of the IEEE conference on Computer Vision and Pattern Recognition (2017) pp. 2001–2010.
  25. D. Rolnick, A. Ahuja, J. Schwarz, T. Lillicrap,  and G. Wayne, “Experience replay for continual learning,” Advances in neural information processing systems 32 (2019).
  26. R. Tiwari, K. Killamsetty, R. Iyer,  and P. Shenoy, “Gcr: Gradient coreset based replay buffer selection for continual learning,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (2022) pp. 99–108.
  27. J. Kirkpatrick, R. Pascanu, N. Rabinowitz, J. Veness, G. Desjardins, A. A. Rusu, K. Milan, J. Quan, T. Ramalho, A. Grabska-Barwinska, et al., “Overcoming catastrophic forgetting in neural networks,” Proceedings of the national academy of sciences 114, 3521–3526 (2017).
  28. G. Bhatt, J. Ross,  and L. Sigal, “Preventing catastrophic forgetting through memory networks in continuous detection,” arXiv preprint arXiv:2403.14797  (2024).
  29. Z. Ke, B. Liu, N. Ma, H. Xu,  and L. Shu, “Achieving forgetting prevention and knowledge transfer in continual learning,” Advances in Neural Information Processing Systems 34, 22443–22456 (2021).
  30. M. M. Bronstein, J. Bruna, Y. LeCun, A. Szlam,  and P. Vandergheynst, “Geometric deep learning: going beyond euclidean data,” IEEE Signal Processing Magazine 34, 18–42 (2017).
  31. O. Ostapenko, M. Puscas, T. Klein, P. Jahnichen,  and M. Nabi, “Learning to remember: A synaptic plasticity driven framework for continual learning,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition (2019) pp. 11321–11329.
  32. Y. Kong, L. Liu, Z. Wang,  and D. Tao, “Balancing stability and plasticity through advanced null space in continual learning,” in European Conference on Computer Vision (Springer, 2022) pp. 219–236.
  33. P. P. Brahma and A. Othon, “Subset replay based continual learning for scalable improvement of autonomous systems,” in 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) (IEEE, 2018) pp. 1179–11798.
  34. I. Loshchilov and F. Hutter, “Sgdr: Stochastic gradient descent with warm restarts,” arXiv preprint arXiv:1608.03983  (2016).
  35. I. Loshchilov and F. Hutter, “Decoupled weight decay regularization,” in International Conference on Learning Representations (2019).
  36. Z. Wang, A. C. Bovik, H. R. Sheikh,  and E. P. Simoncelli, “Image quality assessment: from error visibility to structural similarity,” IEEE transactions on image processing 13, 600–612 (2004).
  37. M. B. Porter and Y.-C. Liu, “Finite-element ray tracing,” Theoretical and computational acoustics 2, 947–956 (1994).
  38. M. B. Porter, “Bellhop: A beam/ray trace code,” https://oalib-acoustics.org/website_resources/AcousticsToolbox/Bellhop-2010-1.pdf (2010).
  39. A. Paszke, S. Gross, F. Massa, A. Lerer, J. Bradbury, G. Chanan, T. Killeen, Z. Lin, N. Gimelshein, L. Antiga, A. Desmaison, A. Kopf, E. Yang, Z. DeVito, M. Raison, A. Tejani, S. Chilamkurthy, B. Steiner, L. Fang, J. Bai,  and S. Chintala, “Pytorch: An imperative style, high-performance deep learning library,” in Advances in Neural Information Processing Systems 32 (Curran Associates, Inc., 2019) pp. 8024–8035.
  40. G. R. Ebbeson and R. G. Turner, “Sound propagation over dickins seamount in the northeast pacific ocean,” The Journal of the Acoustical Society of America 73, 143–152 (1983).
Citations (2)
View on Semantic Scholar

Summary

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

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now
X Twitter Logo Streamline Icon: https://streamlinehq.com

Tweets