paper-with-me

홈 › Papers

Understanding the Impact of Precision Quantization on the Accuracy and Energy of Neural Networks

2016-12-12 · Soheil Hashemi, Nicholas Anthony, Hokchhay Tann, R. Iris Bahar, Sherief Reda

Deep neural networks are gaining in popularity as they are used to generate state-of-the-art results for a variety of computer vision and machine learning applications. At the same time, these networks have grown in depth and complexity in order to solve harder problems. Given the limitations in power budgets dedicated to these networks, the importance of low-power, low-memory solutions has been stressed in recent years. While a large number of dedicated hardware using different precisions has recently been proposed, there exists no comprehensive study of different bit precisions and arithmetic in both inputs and network parameters. In this work, we address this issue and perform a study of different bit-precisions in neural networks (from floating-point to fixed-point, powers of two, and binary). In our evaluation, we consider and analyze the effect of precision scaling on both network accuracy and hardware metrics including memory footprint, power and energy consumption, and design area. We also investigate training-time methodologies to compensate for the reduction in accuracy due to limited bit precision and demonstrate that in most cases, precision scaling can deliver significant benefits in design metrics at the cost of very modest decreases in network accuracy. In addition, we propose that a small portion of the benefits achieved when using lower precisions can be forfeited to increase the network size and therefore the accuracy. We evaluate our experiments, using three well-recognized networks and datasets to show its generality. We investigate the trade-offs and highlight the benefits of using lower precisions in terms of energy and memory footprint.

📄 PDF Abstract BibTeX arXiv:1612.03940

Code (0)

등록된 구현이 없습니다.

Tasks

Quantization

Similar Papers 제목 키워드 기반

FineQ: Software-Hardware Co-Design for Low-Bit Fine-Grained Mixed-Precision Quantization of LLMs

2025-04-28 · Xilong Xie, Liang Wang, Limin Xiao, Meng Han 외

Large language models (LLMs) have significantly advanced the natural language processing paradigm but impose substantial demands on memory and computational resources. Quantization is one of the most effective ways to re…

Quantization

Understanding Efficiency: Quantization, Batching, and Serving Strategies in LLM Energy Use

2026-01-29 · Julien Delavande, Regis Pierrard, Sasha Luccioni arxiv

Large Language Models (LLMs) are increasingly deployed in production, contributing towards shifting the burden in terms of computational resources and energy demands from training to inference. While prior work has exami…

Text Generation

FGMP: Fine-Grained Mixed-Precision Weight and Activation Quantization for Hardware-Accelerated LLM Inference

2025-04-19 · Coleman Hooper, Charbel Sakr, Ben Keller, Rangharajan Venkatesan 외

Quantization is a powerful tool to improve large language model (LLM) inference efficiency by utilizing more energy-efficient low-precision datapaths and reducing memory footprint. However, accurately quantizing LLM weig…

Large Language ModelQuantization

Efficient and Effective Methods for Mixed Precision Neural Network Quantization for Faster, Energy-efficient Inference

2023-01-30 · Deepika Bablani, Jeffrey L. McKinstry, Steven K. Esser, Rathinakumar Appuswamy 외

For efficient neural network inference, it is desirable to achieve state-of-the-art accuracy with the simplest networks requiring the least computation, memory, and power. Quantizing networks to lower precision is a powe…

Efficient Neural NetworkQuantization

Mixed-Precision Quantization for Language Models: Techniques and Prospects

2025-10-19 · Mariam Rakka, Marios Fournarakis, Olga Krestinskaya, Jinane Bazzi 외 arxiv

The rapid scaling of language models (LMs) has resulted in unprecedented computational, memory, and energy requirements, making their training and deployment increasingly unsustainable. Quantization has emerged as an ess…