Deep Polynomial Neural Networks (2006.13026v2)

Abstract: Deep Convolutional Neural Networks (DCNNs) are currently the method of choice both for generative, as well as for discriminative learning in computer vision and machine learning. The success of DCNNs can be attributed to the careful selection of their building blocks (e.g., residual blocks, rectifiers, sophisticated normalization schemes, to mention but a few). In this paper, we propose $\Pi$-Nets, a new class of function approximators based on polynomial expansions. $\Pi$-Nets are polynomial neural networks, i.e., the output is a high-order polynomial of the input. The unknown parameters, which are naturally represented by high-order tensors, are estimated through a collective tensor factorization with factors sharing. We introduce three tensor decompositions that significantly reduce the number of parameters and show how they can be efficiently implemented by hierarchical neural networks. We empirically demonstrate that $\Pi$-Nets are very expressive and they even produce good results without the use of non-linear activation functions in a large battery of tasks and signals, i.e., images, graphs, and audio. When used in conjunction with activation functions, $\Pi$-Nets produce state-of-the-art results in three challenging tasks, i.e. image generation, face verification and 3D mesh representation learning. The source code is available at \url{https://github.com/grigorisg9gr/polynomial_nets}.

• The paper's primary contribution is unclear due to the use of placeholder text in its provided template.
• It appears to aim at integrating polynomial functions into deep neural network architectures for enhanced nonlinear modeling.
• Key methodologies and experimental evaluations are not detailed, leaving actual performance and implementation insights undetermined.

Thank you for providing the document. It appears that the content you've provided is the LaTeX source code for a template file (IEEEtran.cls bare_adv.tex) used for preparing papers for IEEE Computer Society journals, not the actual content of the paper "Deep Polynomial Neural Networks" (2006.13026).

The file contains placeholder text such as "The abstract goes here", "Introduction text here", and "The conclusion goes here". Therefore, I cannot provide a summary of the technical contributions, implementation details, or practical applications of the "Deep Polynomial Neural Networks" paper based on this template file.

If you can provide the actual content of the paper "Deep Polynomial Neural Networks" (2006.13026), I would be happy to analyze it and discuss its practical implementation and application.