Fixed-point Factorized Networks
This work addresses the resource constraints of embedded systems like smartphones by making deep neural networks more efficient, though it is incremental as it builds on existing pretrained models.
The paper tackles the problem of deep neural networks being too computationally intensive for embedded systems by introducing Fixed-point Factorized Networks (FFN), which reduce computational complexity and storage by using weights of -1, 0, and 1, achieving comparable accuracy with only one-thousandth of the multiply operations on ImageNet.
In recent years, Deep Neural Networks (DNN) based methods have achieved remarkable performance in a wide range of tasks and have been among the most powerful and widely used techniques in computer vision. However, DNN-based methods are both computational-intensive and resource-consuming, which hinders the application of these methods on embedded systems like smart phones. To alleviate this problem, we introduce a novel Fixed-point Factorized Networks (FFN) for pretrained models to reduce the computational complexity as well as the storage requirement of networks. The resulting networks have only weights of -1, 0 and 1, which significantly eliminates the most resource-consuming multiply-accumulate operations (MACs). Extensive experiments on large-scale ImageNet classification task show the proposed FFN only requires one-thousandth of multiply operations with comparable accuracy.