Deep Learning with Limited Numerical Precision
This addresses computational efficiency for deep learning practitioners, offering an incremental improvement in hardware acceleration.
The paper tackles the problem of training large-scale deep neural networks under computational constraints by exploring limited precision data representation, showing that deep networks can be trained using only 16-bit fixed-point numbers with stochastic rounding, incurring little to no accuracy degradation.
Training of large-scale deep neural networks is often constrained by the available computational resources. We study the effect of limited precision data representation and computation on neural network training. Within the context of low-precision fixed-point computations, we observe the rounding scheme to play a crucial role in determining the network's behavior during training. Our results show that deep networks can be trained using only 16-bit wide fixed-point number representation when using stochastic rounding, and incur little to no degradation in the classification accuracy. We also demonstrate an energy-efficient hardware accelerator that implements low-precision fixed-point arithmetic with stochastic rounding.