Bayesian Sparsification Methods for Deep Complex-valued Networks
This work addresses compression for complex-valued networks in embedded applications, but it is incremental as it adapts an existing method to a new domain.
The authors tackled the problem of compressing complex-valued neural networks for embedded systems by extending Sparse Variational Dropout to complex domains, achieving a 50-100x compression on MusicNet with a small performance penalty while replicating state-of-the-art results.
With continual miniaturization ever more applications of deep learning can be found in embedded systems, where it is common to encounter data with natural complex domain representation. To this end we extend Sparse Variational Dropout to complex-valued neural networks and verify the proposed Bayesian technique by conducting a large numerical study of the performance-compression trade-off of C-valued networks on two tasks: image recognition on MNIST-like and CIFAR10 datasets and music transcription on MusicNet. We replicate the state-of-the-art result by Trabelsi et al. [2018] on MusicNet with a complex-valued network compressed by 50-100x at a small performance penalty.