paper-with-me

홈 › Papers

A Metric Driven Approach to Mixed Precision Training

2024-08-06 · Mitchelle Rasquinha, Gil Tabak

As deep learning methodologies have developed, it has been generally agreed that increasing neural network size improves model quality. However, this is at the expense of memory and compute requirements, which also need to be increased. Various efficiency techniques have been proposed to rein in hardware costs, one being the use of low precision numerics. Recent accelerators have introduced several different 8-bit data types to help accommodate DNNs in terms of numerics. In this paper, we identify a metric driven methodology to aid in the choice of numerics. We demonstrate how such a methodology can help scale training of a language representation model. The technique can be generalized to other model architectures.

📄 PDF Abstract BibTeX arXiv:2408.02897

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

BMPQ: Bit-Gradient Sensitivity Driven Mixed-Precision Quantization of DNNs from Scratch

2021-12-24 · Souvik Kundu, Shikai Wang, Qirui Sun, Peter A. Beerel 외

Large DNNs with mixed-precision quantization can achieve ultra-high compression while retaining high classification performance. However, because of the challenges in finding an accurate metric that can guide the optimiz…

QuantizationSensitivity

Mix-QViT: Mixed-Precision Vision Transformer Quantization Driven by Layer Importance and Quantization Sensitivity

2025-01-10 · Navin Ranjan, Andreas Savakis

In this paper, we propose Mix-QViT, an explainability-driven MPQ framework that systematically allocates bit-widths to each layer based on two criteria: layer importance, assessed via Layer-wise Relevance Propagation (LR…

QuantizationSensitivity

Value-Driven Mixed-Precision Quantization for Patch-Based Inference on Microcontrollers

2024-01-24 · Wei Tao, Shenglin He, Kai Lu, Xiaoyang Qu 외

Deploying neural networks on microcontroller units (MCUs) presents substantial challenges due to their constrained computation and memory resources. Previous researches have explored patch-based inference as a strategy t…

Quantization

SNIP: An Adaptive Mixed Precision Framework for Subbyte Large Language Model Training

2026-02-01 · Yunjie Pan, Yongyi Yang, Hanmei Yang, Scott Mahlke arxiv

Training large language models (LLMs) efficiently while preserving model quality poses significant challenges, particularly with subbyte precision supported by state-of-the-art GPUs. Current mixed-precision training appr…

Mixed-Precision Conjugate Gradient Solvers with RL-Driven Precision Tuning

2025-04-19 · Xinye Chen

This paper presents a novel reinforcement learning (RL) framework for dynamically optimizing numerical precision in the preconditioned conjugate gradient (CG) method. By modeling precision selection as a Markov Decision …

Computational EfficiencyQ-LearningReinforcement Learning (RL)