A Dimensionality Reduction Approach for Convolutional Neural Networks
This work addresses storage constraints in embedded systems, but it is incremental as it adapts existing reduction methods to neural networks.
The paper tackles the problem of compressing Convolutional Neural Networks for embedded systems by applying classical dimensionality reduction techniques like Active Subspaces and Proper Orthogonal Decomposition to reduce layers, achieving similar accuracy with memory savings.
The focus of this paper is the application of classical model order reduction techniques, such as Active Subspaces and Proper Orthogonal Decomposition, to Deep Neural Networks. We propose a generic methodology to reduce the number of layers of a pre-trained network by combining the aforementioned techniques for dimensionality reduction with input-output mappings, such as Polynomial Chaos Expansion and Feedforward Neural Networks. The necessity of compressing the architecture of an existing Convolutional Neural Network is motivated by its application in embedded systems with specific storage constraints. Our experiment shows that the reduced nets obtained can achieve a level of accuracy similar to the original Convolutional Neural Network under examination, while saving in memory allocation.