A compressive sensing based compressed neural network for sound source localization

Microphone arrays are today employed to specify the sound source locations in numerous real time applications such as speech processing in large rooms or acoustic echo cancellation. Signal sources may exist in the near field or far field with respect to the microphones. Current Neural Networks (NNs) based source localization approaches assume far field narrowband sources. One of the important limitations of these NN-based approaches is making balance between computational complexity and the development of NNs; an architecture that is too large or too small will affect the performance in terms of generalization and computational cost. In the previous analysis, saliency subject has been employed to determine the most suitable structure, however, it is time-consuming and the performance is not robust. In this paper, a family of new algorithms for compression of NNs is presented based on Compressive Sampling (CS) theory. The proposed framework makes it possible to find a sparse structure for NNs, and then the designed neural network is compressed by using CS. The key difference between our algorithm and the state-of-the-art techniques is that the mapping is continuously done using the most effective features; therefore, the proposed method has a fast convergence. The empirical work demonstrates that the proposed algorithm is an effective alternative to traditional methods in terms of accuracy and computational complexity.

[1]  Emmanuel J. Candès,et al.  Robust uncertainty principles: exact signal reconstruction from highly incomplete frequency information , 2004, IEEE Transactions on Information Theory.

[2]  Robert D. Nowak,et al.  Signal Reconstruction From Noisy Random Projections , 2006, IEEE Transactions on Information Theory.

[3]  Pierre Vandergheynst,et al.  Compressed Sensing and Redundant Dictionaries , 2007, IEEE Transactions on Information Theory.

[4]  Jocelyn Sietsma,et al.  Creating artificial neural networks that generalize , 1991, Neural Networks.

[5]  Tao Xu,et al.  A compressed sensing approach for underdetermined blind audio source separation with sparse representation , 2009, 2009 IEEE/SP 15th Workshop on Statistical Signal Processing.

[6]  Pierre Vandergheynst,et al.  Compressed Sensing: “When Sparsity Meets Sampling” , 2011 .

[7]  F. A. Sakarya,et al.  Speaker localization for far-field and near-field wideband sources using neural networks , 1999, NSIP.

[8]  Babak Hassibi,et al.  Second Order Derivatives for Network Pruning: Optimal Brain Surgeon , 1992, NIPS.

[9]  Russell Reed,et al.  Pruning algorithms-a survey , 1993, IEEE Trans. Neural Networks.

[10]  Ai-Jun Li,et al.  Information geometry on pruning of neural network , 2004, Proceedings of 2004 International Conference on Machine Learning and Cybernetics (IEEE Cat. No.04EX826).

[11]  Alfred Jean Philippe Lauret,et al.  A node pruning algorithm based on a Fourier amplitude sensitivity test method , 2006, IEEE Transactions on Neural Networks.

[12]  Masafumi Hagiwara,et al.  Removal of hidden units and weights for back propagation networks , 1993, Proceedings of 1993 International Conference on Neural Networks (IJCNN-93-Nagoya, Japan).

[13]  Yann LeCun,et al.  Optimal Brain Damage , 1989, NIPS.

[14]  Abdesselam Bouzerdoum,et al.  A Neural Network pruning approach based on Compressive Sampling , 2009, 2009 International Joint Conference on Neural Networks.