Tensor-based framework for training flexible neural networks
This work addresses model compression for neural networks, offering an incremental improvement by optimizing activation functions to reduce model size.
The paper tackled the problem of compressing large pretrained neural networks by proposing a tensor-based framework to estimate flexible activation functions as weighted sums of basis functions, achieving compression by replacing subnetworks with new flexible layers, as demonstrated on a pretrained CNN for character classification.
Activation functions (AFs) are an important part of the design of neural networks (NNs), and their choice plays a predominant role in the performance of a NN. In this work, we are particularly interested in the estimation of flexible activation functions using tensor-based solutions, where the AFs are expressed as a weighted sum of predefined basis functions. To do so, we propose a new learning algorithm which solves a constrained coupled matrix-tensor factorization (CMTF) problem. This technique fuses the first and zeroth order information of the NN, where the first-order information is contained in a Jacobian tensor, following a constrained canonical polyadic decomposition (CPD). The proposed algorithm can handle different decomposition bases. The goal of this method is to compress large pretrained NN models, by replacing subnetworks, {\em i.e.,} one or multiple layers of the original network, by a new flexible layer. The approach is applied to a pretrained convolutional neural network (CNN) used for character classification.