Adaptive Precision Training: Quantify Back Propagation in Neural Networks with Fixed-point Numbers
arXiv:1911.00361
Abstract
Adaptive Precision Training: Quantify Back Propagation in Neural Networks with Fixed-point Numbers. Recent emerged quantization technique has been applied to inference of deep neural networks for fast and efficient execution. However, directly applying quantization in training can cause significant accuracy loss, thus remaining an open challenge.
We would like to withdraw the manuscript because it lacks of comparisons. The main contribution is not well verified by experiments
References in corpus (8)
- Very Deep Convolutional Networks for Large-Scale Image Recognition
- MXNet: A Flexible and Efficient Machine Learning Library for Heterogeneous Distributed Systems
- DoReFa-Net: Training Low Bitwidth Convolutional Neural Networks with Low Bitwidth Gradients
- PACT: Parameterized Clipping Activation for Quantized Neural Networks
- Fixed Point Quantization of Deep Convolutional Networks
- Training Deep Neural Networks with 8-bit Floating Point Numbers
- Mixed Precision Quantization of ConvNets via Differentiable Neural Architecture Search
- Sockeye: A Toolkit for Neural Machine Translation