Performance-Efficiency Trade-off of Low-Precision Numerical Formats in Deep Neural Networks
This work addresses the challenge of deploying efficient DNNs on edge-devices by optimizing inference time and power consumption, representing an incremental improvement in low-precision computing methods.
The study investigated the performance-efficiency trade-offs of low-precision numerical formats (fixed-point, floating point, and posit) at ≤8-bit precision in deep neural network inference, finding that posits outperform other formats at this precision with competitive resource requirements.
Deep neural networks (DNNs) have been demonstrated as effective prognostic models across various domains, e.g. natural language processing, computer vision, and genomics. However, modern-day DNNs demand high compute and memory storage for executing any reasonably complex task. To optimize the inference time and alleviate the power consumption of these networks, DNN accelerators with low-precision representations of data and DNN parameters are being actively studied. An interesting research question is in how low-precision networks can be ported to edge-devices with similar performance as high-precision networks. In this work, we employ the fixed-point, floating point, and posit numerical formats at $\leq$8-bit precision within a DNN accelerator, Deep Positron, with exact multiply-and-accumulate (EMAC) units for inference. A unified analysis quantifies the trade-offs between overall network efficiency and performance across five classification tasks. Our results indicate that posits are a natural fit for DNN inference, outperforming at $\leq$8-bit precision, and can be realized with competitive resource requirements relative to those of floating point.