Doctor of Crosswise: Reducing Over-parametrization in Neural Networks
arXiv:1905.10324
Abstract
Dr. of Crosswise proposes a new architecture to reduce over-parametrization in Neural Networks. It introduces an operand for rapid computation in the framework of Deep Learning that leverages learned weights. The formalism is described in detail providing both an accurate elucidation of the mechanics and the theoretical implications.
References in corpus (8)
- Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
- Distributed Representations of Words and Phrases and their Compositionality
- Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks
- Learning a Probabilistic Latent Space of Object Shapes via 3D Generative-Adversarial Modeling
- AutoAugment: Learning Augmentation Policies from Data
- High-resolution Deep Convolutional Generative Adversarial Networks
- A la Carte - Learning Fast Kernels
- McKernel: A Library for Approximate Kernel Expansions in Log-linear Time