Recurrent Neural Networks With Limited Numerical Precision
This addresses resource efficiency for deploying RNNs on specialized low-power hardware, though it is incremental as it applies existing quantization methods to RNNs.
The paper tackled the problem of high computational demands in Recurrent Neural Networks (RNNs) by limiting numerical precision of weights and biases, resulting in low-precision RNNs that achieved similar or higher accuracy on certain datasets.
Recurrent Neural Networks (RNNs) produce state-of-art performance on many machine learning tasks but their demand on resources in terms of memory and computational power are often high. Therefore, there is a great interest in optimizing the computations performed with these models especially when considering development of specialized low-power hardware for deep networks. One way of reducing the computational needs is to limit the numerical precision of the network weights and biases, and this will be addressed for the case of RNNs. We present results from the use of different stochastic and deterministic reduced precision training methods applied to two major RNN types, which are then tested on three datasets. The results show that the stochastic and deterministic ternarization, pow2- ternarization, and exponential quantization methods gave rise to low-precision RNNs that produce similar and even higher accuracy on certain datasets, therefore providing a path towards training more efficient implementations of RNNs in specialized hardware.