Algorithm-Hardware Co-Design of Distribution-Aware Logarithmic-Posit Encodings for Efficient DNN Inference
Traditional Deep Neural Network (DNN) quantization methods using integer, fixed-point, or floating-point data types struggle to capture diverse DNN parameter distributions at low precision, and often require large silicon overhead and intensive quantization-aware training. In this study, we introduce Logarithmic Posits (LP), an adaptive, hardware-friendly data type inspired by posits that dynamically adapts to DNN weight/activation distributions by parameterizing LP bit fields. We also develop a novel genetic-algorithm based framework, LP Quantization (LPQ), to find optimal layer-wise LP parameters while reducing representational divergence between quantized and full-precision models through a novel global-local contrastive objective. Additionally, we design a unified mixed-precision LP accelerator (LPA) architecture comprising of processing elements (PEs) incorporating LP in the computational datapath. Our algorithm-hardware co-design demonstrates on average <1% drop in top-1 accuracy across various CNN and ViT models. It also achieves ~ 2x improvements in performance per unit area and 2.2x gains in energy efficiency compared to state-of-the-art quantization accelerators using different data types.
Code (1)
Tasks
QuantizationSimilar Papers 제목 키워드 기반
Power-of-Two Quantization for Low Bitwidth and Hardware Compliant Neural Networks
Deploying Deep Neural Networks in low-power embedded devices for real time-constrained applications requires optimization of memory and computational complexity of the networks, usually by quantizing the weights. Most of…
QuantizationHardware Complexity Aware Design Strategy for a Fused Logarithmic and Anti-Logarithmic Converter
The logarithmic and anti-logarithmic converters are realized with the piecewise linear approximation method, which is implemented by the shift-and-add architecture. This brief utilizes the similarities of Log and Antilog…
Automated Log-Scale Quantization for Low-Cost Deep Neural Networks
Quantization plays an important role in deep neural network (DNN) hardware. In particular, logarithmic quantization has multiple advantages for DNN hardware implementations, and its weakness in terms of lower perform…
Image EnhancementQuantizationSemantic SegmentationA Time-to-first-spike Coding and Conversion Aware Training for Energy-Efficient Deep Spiking Neural Network Processor Design
In this paper, we present an energy-efficient SNN architecture, which can seamlessly run deep spiking neural networks (SNNs) with improved accuracy. First, we propose a conversion aware training (CAT) to reduce ANN-to-SN…
Bandit algorithms: Letting go of logarithmic regret for statistical robustness
We study regret minimization in a stochastic multi-armed bandit setting and establish a fundamental trade-off between the regret suffered under an algorithm, and its statistical robustness. Considering broad classes of u…