BLoad: Enhancing Neural Network Training with Efficient Sequential Data Handling

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Ruschel, Raphael, Iftekhar, A. S. M., Manjunath, B. S., You, Suya
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910423625760768
author Ruschel, Raphael
Iftekhar, A. S. M.
Manjunath, B. S.
You, Suya
author_facet Ruschel, Raphael
Iftekhar, A. S. M.
Manjunath, B. S.
You, Suya
contents The increasing complexity of modern deep neural network models and the expanding sizes of datasets necessitate the development of optimized and scalable training methods. In this white paper, we addressed the challenge of efficiently training neural network models using sequences of varying sizes. To address this challenge, we propose a novel training scheme that enables efficient distributed data-parallel training on sequences of different sizes with minimal overhead. By using this scheme we were able to reduce the padding amount by more than 100$x$ while not deleting a single frame, resulting in an overall increased performance on both training time and Recall in our experiments.
format Preprint
id arxiv_https___arxiv_org_abs_2310_10879
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle BLoad: Enhancing Neural Network Training with Efficient Sequential Data Handling
Ruschel, Raphael
Iftekhar, A. S. M.
Manjunath, B. S.
You, Suya
Machine Learning
Distributed, Parallel, and Cluster Computing
The increasing complexity of modern deep neural network models and the expanding sizes of datasets necessitate the development of optimized and scalable training methods. In this white paper, we addressed the challenge of efficiently training neural network models using sequences of varying sizes. To address this challenge, we propose a novel training scheme that enables efficient distributed data-parallel training on sequences of different sizes with minimal overhead. By using this scheme we were able to reduce the padding amount by more than 100$x$ while not deleting a single frame, resulting in an overall increased performance on both training time and Recall in our experiments.
title BLoad: Enhancing Neural Network Training with Efficient Sequential Data Handling
topic Machine Learning
Distributed, Parallel, and Cluster Computing
url https://arxiv.org/abs/2310.10879