2000 character limit reached
Neural Harmonium: An Interpretable Deep Structure for Nonlinear Dynamic System Identification with Application to Audio Processing
Published 10 Oct 2023 in cs.SD, cs.LG, cs.SY, eess.AS, and eess.SY | (2310.07032v1)
Abstract: Improving the interpretability of deep neural networks has recently gained increased attention, especially when the power of deep learning is leveraged to solve problems in physics. Interpretability helps us understand a model's ability to generalize and reveal its limitations. In this paper, we introduce a causal interpretable deep structure for modeling dynamic systems. Our proposed model makes use of the harmonic analysis by modeling the system in a time-frequency domain while maintaining high temporal and spectral resolution. Moreover, the model is built in an order recursive manner which allows for fast, robust, and exact second order optimization without the need for an explicit Hessian calculation. To circumvent the resulting high dimensionality of the building blocks of our system, a neural network is designed to identify the frequency interdependencies. The proposed model is illustrated and validated on nonlinear system identification problems as required for audio signal processing tasks. Crowd-sourced experimentation contrasting the performance of the proposed approach to other state-of-the-art solutions on an acoustic echo cancellation scenario confirms the effectiveness of our method for real-life applications.
- Advances in Network and Acoustic Echo Cancellation. Digital Signal Processing. Springer Berlin Heidelberg. ISBN 9783662044377.
- A new class of wavelet networks for nonlinear system identification. IEEE Transactions on neural networks, 16(4): 862–874.
- Bouc, R. 1971. Modèle mathématique d’hystérésis: application aux systèmes à un degré de liberté. In Acustica, 16–25.
- Inverse scattering for discrete transmission-line models. SIAM review, 29(3): 359–389.
- Applied optimal control: optimization, estimation, and control. Routledge.
- Blind Signal Processing for Time-varying Convolutive Mixing Systems Based on Sequence Estimation on Partly Smooth Manifolds. In 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 7913–7917.
- Volterra-series-based nonlinear system modeling and its engineering applications: A state-of-the-art review. Mechanical Systems and Signal Processing, 87: 340–364.
- Fast, recursive-least-squares transversal filters for adaptive filtering. IEEE Transactions on Acoustics, Speech, and Signal Processing, 32(2): 304–337.
- Cremer, L. 1981. Physik der Geige. Hirzel Stuttgart.
- ICASSP 2022 Acoustic Echo Cancellation Challenge. In ICASSP 2022. The dataset uses other datasets licensed under CC0 license: https://research.google.com/audioset/index.html, https://freesound.org/, and https://zenodo.org/record/1227121#.XRKKxYhKiUk.
- Dobrucki, A. 2011. Nonlinear distortions in electroacoustic devices. Archives of Acoustics, 36(2): 437–460.
- Considerations for evaluation and generalization in interpretable machine learning. In Explainable and interpretable models in computer vision and machine learning, 3–17. Springer.
- Haykin, S. 2002. Adaptive filter theory. Upper Saddle River, NJ: Prentice Hall, 4th edition.
- Time-Frequency Analysis. ISTE. Wiley. ISBN 9781118623831.
- Kalman Filtering with Newton’s Method [Lecture Notes]. IEEE Control Systems Magazine, 30(6): 101–106.
- ITU-T. 1996. Recommendation P.800: Methods for subjective determination of transmission quality.
- ITU-T. 1998. Recommendation P.831: Subjective performance evaluation of network echo cancellers.
- Adam: A method for stochastic optimization. In Proceedings of the 3rd International Conference Learning Representations.
- Cascaded tuning to amplitude modulation for natural sound recognition. Journal of Neuroscience, 39(28): 5517–5533.
- Topmoumoute Online Natural Gradient Algorithm. In NIPS, 849–856. Citeseer.
- Lecca, P. 2021. Machine learning for causal inference in biological networks: perspectives of this challenge. Frontiers in Bioinformatics, 45.
- Fourier Neural Operator for Parametric Partial Differential Equations. arXiv:2010.08895.
- Optimizing neural networks with kronecker-factored approximate curvature. In International conference on machine learning, 2408–2417. PMLR.
- Modified MP3 encoder using complex modified cosine transform. In 2003 International Conference on Multimedia and Expo. ICME ’03. Proceedings, volume 2, II–709.
- System identification using Bayesian neural networks with nonparametric noise models. arXiv preprint arXiv:2104.12119.
- Molnar, C. 2018. Interprtable machine learning: A guide for making black box models explainable.
- Müller, S. 1999. Digitale Signalverarbeitung für Lautsprecher. Ph.D. thesis, Rheinisch-Westfälische Technische Hochschule Aachen, Aachen.
- Definitions, methods, and applications in interpretable machine learning. Proceedings of the National Academy of Sciences, 116(44): 22071–22080.
- Ollivier, Y. 2015. Riemannian metrics for neural networks I: feedforward networks. Information and Inference: A Journal of the IMA, 4(2): 108–153.
- Discrete-Time Signal Processing. Prentice Hall, 2nd edition. ISBN 0137549202.
- PESQ. 2001. ITU-T, P.862: Perceptual evaluation of speech quality (PESQ): An objective method for end-to-end speech quality assessment of narrow-band telephone networks and speech codecs.
- Kernel adaptive filtering: a comprehensive introduction. John Wiley & Sons.
- Ranking of the nonlinearities of electrodynamic loudspeakers. Archives of Acoustics, 35(1): 49–66.
- Explainable artificial intelligence: understanding, visualizing and interpreting deep learning models. ITU Journal: ICT Discoveries 1 (1), 39–48.
- Smolensky, P. 1986. Information processing in dynamical systems: Foundations of harmony theory. Technical report, Colorado Univ at Boulder Dept of Computer Science.
- Modeling and real-time auralization of electrodynamic loudspeaker non-linearities. In 2004 IEEE International Conference on Acoustics, Speech, and Signal Processing, volume 4, iv–iv. IEEE.
- ICASSP 2021 Acoustic Echo Cancellation Challenge: Datasets, Testing Framework, and Results.
- Digital sound synthesis by physical modeling using the functional transformation method. Springer Science & Business Media.
- Low-Complexity, Real-Time Joint Neural Echo Control and Speech Enhancement Based On PercepNet.
- von Helmholtz, H. 1898. Vorlesungen über die mathematischen Principien der Akustik, volume 3. JA Barth.
- Waldron, S. F. 2018. An introduction to finite tight frames. Springer.
- WebRTC. 2022. The WebRTC Code Repository.
- Wen, Y.-K. 1976. Method for random vibration of hysteretic systems. Journal of the engineering mechanics division, 102(2): 249–263.
- Identification of hammerstein–wiener models. Automatica, 49(1): 70–81.
- Linear and nonlinear inverse scattering. SIAM Journal on Applied Mathematics, 59(2): 669–699.
- Zhang, Q. 1997. Using wavelet network in nonparametric estimation. IEEE Transactions on Neural networks, 8(2): 227–236.
Paper Prompts
Sign up for free to create and run prompts on this paper using GPT-5.
Top Community Prompts
Collections
Sign up for free to add this paper to one or more collections.
