A Dimensionality Reduction Approach for Convolutional Neural Networks
arXiv:2110.09163 · doi:10.1007/s10489-023-04730-1
Abstract
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.
References in corpus (22)
- Very Deep Convolutional Networks for Large-Scale Image Recognition
- SciPy 1.0--Fundamental Algorithms for Scientific Computing in Python
- PyTorch: An Imperative Style, High-Performance Deep Learning Library
- Distilling the Knowledge in a Neural Network
- Knowledge Distillation: A Survey
- A Survey of the Recent Architectures of Deep Convolutional Neural Networks
- Natural Language Processing: State of The Art, Current Trends and Challenges
- Recent Trends in Deep Learning Based Natural Language Processing
- Rethinking the Value of Network Pruning
- Binarized Neural Networks
- Tensorizing Neural Networks
- Active subspace methods in theory and practice: applications to kriging surfaces
- Low-Rank Tensor Networks for Dimensionality Reduction and Large-Scale Optimization Problems: Perspectives and Challenges PART 1
- Tensor Networks for Dimensionality Reduction and Large-Scale Optimizations. Part 2 Applications and Future Perspectives
- Provable approximation properties for deep neural networks
- Deep Neural Network Approximation for Custom Hardware: Where We've Been, Where We're Going
- Advances in Reduced Order Methods for Parametric Industrial Problems in Computational Fluid Dynamics
- Resource-Efficient Neural Networks for Embedded Systems
- ATHENA: Advanced Techniques for High Dimensional Parameter Spaces to Enhance Numerical Analysis
- Kernel-based active subspaces with application to computational fluid dynamics parametric problems using the discontinuous Galerkin method
- Model Order Reduction in Neuroscience
- A complete data-driven framework for the efficient solution of parametric shape design and optimisation in naval engineering problems