This work addresses the design challenges of acoustic metamaterials for noise attenuation, focusing on Multiresonant Layered Acoustic Metamaterial (MLAM) panels that excel at attenuating low-frequency noise below 1000 Hz [1]. By leveraging the coupling effect between two resonant frequencies, these panels achieve substantial noise attenuation compared to equivalent mass homogeneous materials. The proposed two-phase design strategy integrates a neural network-based surrogate model for efficient assessment of individual effective layer properties, computed through homogenization, and a topology optimization approach to maximize noise attenuation while preserving a coupled Sound Transmission Loss (STL) response. The optimization process, guided by genetic algorithms in combination with DNN-based models, results in configurations with noise attenuation exceeding 20 dB over 330 Hz in comparison to homogeneous materials with the same surface density. Notably, the use of the surrogate model enhances accuracy in predicting effective properties and evaluating STL, outperforming conventional polynomial interpolation techniques. Furthermore, the ML-based model substantially reduces the computational cost of optimizations in several order of magnitude compared to direct numerical simulations. This cost-effective approach enables the optimization of multiple objective functions and constraints. [2,3] [1] D. Roca, J. Cante, O. Lloberas-Valls, T. Pàmies, and J. Oliver. Multiresonant Layered Acoustic Metamaterial (MLAM) solution for broadband low-frequency noise attenuation through double-peak sound transmission loss response. Extreme Mechanics Letters, Vol. 47, p. 101368, Aug. 2021. DOI: 10.1016/j.eml.2021.101368. [2] G. Sal-Anglada, D. Yago, J. Cante, J. Oliver, and D. Roca. Optimal design of Multiresonant Layered Acoustic Metamaterials (MLAM) via a homogenization approach. Engineering Structures, Vol. 293, p. 116555, Oct. 2023. DOI: 10.1016/j.engstruct.2023.116555. [3] D. Yago, G. Sal-Anglada, D. Roca, J. Cante, and J. Oliver. Machine learning in solid mechanics: application to acoustic metamaterial design. International Journal for Numerical Methods in Engineering. Under Review.