Deep Adaptive Wavelet Network
This addresses the problem of cumbersome trial-and-error design and interpretability issues in computer vision for researchers and practitioners, though it is incremental as it builds on existing wavelet methods.
The paper tackles the lack of interpretability and manual design in convolutional neural networks by proposing a deep neural network that integrates multiresolution analysis, achieving competitive accuracy in image classification tasks with less hyper-parameter tuning.
Even though convolutional neural networks have become the method of choice in many fields of computer vision, they still lack interpretability and are usually designed manually in a cumbersome trial-and-error process. This paper aims at overcoming those limitations by proposing a deep neural network, which is designed in a systematic fashion and is interpretable, by integrating multiresolution analysis at the core of the deep neural network design. By using the lifting scheme, it is possible to generate a wavelet representation and design a network capable of learning wavelet coefficients in an end-to-end form. Compared to state-of-the-art architectures, the proposed model requires less hyper-parameter tuning and achieves competitive accuracy in image classification tasks