paper-with-me

Papers

A Study of BFLOAT16 for Deep Learning Training

2019-05-29 · Dhiraj Kalamkar, Dheevatsa Mudigere, Naveen Mellempudi, Dipankar Das, Kunal Banerjee, Sasikanth Avancha, Dharma Teja Vooturi, Nataraj Jammalamadaka, Jianyu Huang, Hector Yuen, Jiyan Yang, Jongsoo Park, Alexander Heinecke, Evangelos Georganas, Sudarshan Srinivasan, Abhisek Kundu, Misha Smelyanskiy, Bharat Kaul, Pradeep Dubey

This paper presents the first comprehensive empirical study demonstrating the efficacy of the Brain Floating Point (BFLOAT16) half-precision format for Deep Learning training across image classification, speech recognition, language modeling, generative networks and industrial recommendation systems. BFLOAT16 is attractive for Deep Learning training for two reasons: the range of values it can represent is the same as that of IEEE 754 floating-point format (FP32) and conversion to/from FP32 is simple. Maintaining the same range as FP32 is important to ensure that no hyper-parameter tuning is required for convergence; e.g., IEEE 754 compliant half-precision floating point (FP16) requires hyper-parameter tuning. In this paper, we discuss the flow of tensors and various key operations in mixed precision training, and delve into details of operations, such as the rounding modes for converting FP32 tensors to BFLOAT16. We have implemented a method to emulate BFLOAT16 operations in Tensorflow, Caffe2, IntelCaffe, and Neon for our experiments. Our results show that deep learning training using BFLOAT16 tensors achieves the same state-of-the-art (SOTA) results across domains as FP32 tensors in the same number of iterations and with no changes to hyper-parameters.

📄 PDF Abstract BibTeX arXiv:1905.12322

Code (0)

등록된 구현이 없습니다.

Tasks

Deep Learningimage-classificationImage ClassificationLanguage ModelingLanguage ModellingRecommendation Systemsspeech-recognitionSpeech Recognition

Similar Papers 제목 키워드 기반

When Precision Meets Position: BFloat16 Breaks Down RoPE in Long-Context Training

2024-11-20 · Haonan Wang, Qian Liu, Chao Du, Tongyao Zhu 외

Extending context window sizes allows large language models (LLMs) to process longer sequences and handle more complex tasks. Rotary Positional Embedding (RoPE) has become the de facto standard due to its relative positi…

Computational EfficiencyPosition

The Effects of Approximate Multiplication on Convolutional Neural Networks

2020-07-20 · Min Soo Kim, Alberto A. Del Barrio, HyunJin Kim, Nader Bagherzadeh

This paper analyzes the effects of approximate multiplication when performing inferences on deep convolutional neural networks (CNNs). The approximate multiplication can reduce the cost of the underlying circuits so that…

Revisiting BFloat16 Training

2020-10-13 · Pedram Zamirai, Jian Zhang, Christopher R. Aberger, Christopher De Sa

State-of-the-art generic low-precision training algorithms use a mix of 16-bit and 32-bit precision, creating the folklore that 16-bit hardware compute units alone are not enough to maximize model accuracy. As a result, …

Deep Learning

Compressed Real Numbers for AI: a case-study using a RISC-V CPU

2023-09-11 · Federico Rossi, Marco Cococcioni, Roger Ferrer Ibàñez, Jesùs Labarta 외

As recently demonstrated, Deep Neural Networks (DNN), usually trained using single precision IEEE 754 floating point numbers (binary32), can also work using lower precision. Therefore, 16-bit and 8-bit compressed format …

CPU

Representation range needs for 16-bit neural network training

2021-03-29 · Valentina Popescu, Abhinav Venigalla, Di wu, Robert Schreiber

Deep learning has grown rapidly thanks to its state-of-the-art performance across a wide range of real-world applications. While neural networks have been trained using IEEE-754 binary32 arithmetic, the rapid growth of c…