Saved in:
Bibliographic Details
Main Authors: Awobade, Busayo, Oduwole, Mardiyyah, Kolawole, Steven
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2404.04759
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914743557554176
author Awobade, Busayo
Oduwole, Mardiyyah
Kolawole, Steven
author_facet Awobade, Busayo
Oduwole, Mardiyyah
Kolawole, Steven
contents Compression techniques have been crucial in advancing machine learning by enabling efficient training and deployment of large-scale language models. However, these techniques have received limited attention in the context of low-resource language models, which are trained on even smaller amounts of data and under computational constraints, a scenario known as the "low-resource double-bind." This paper investigates the effectiveness of pruning, knowledge distillation, and quantization on an exclusively low-resourced, small-data language model, AfriBERTa. Through a battery of experiments, we assess the effects of compression on performance across several metrics beyond accuracy. Our study provides evidence that compression techniques significantly improve the efficiency and effectiveness of small-data language models, confirming that the prevailing beliefs regarding the effects of compression on large, heavily parameterized models hold true for less-parameterized, small-data models.
format Preprint
id arxiv_https___arxiv_org_abs_2404_04759
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle What Happens When Small Is Made Smaller? Exploring the Impact of Compression on Small Data Pretrained Language Models
Awobade, Busayo
Oduwole, Mardiyyah
Kolawole, Steven
Computation and Language
Machine Learning
Compression techniques have been crucial in advancing machine learning by enabling efficient training and deployment of large-scale language models. However, these techniques have received limited attention in the context of low-resource language models, which are trained on even smaller amounts of data and under computational constraints, a scenario known as the "low-resource double-bind." This paper investigates the effectiveness of pruning, knowledge distillation, and quantization on an exclusively low-resourced, small-data language model, AfriBERTa. Through a battery of experiments, we assess the effects of compression on performance across several metrics beyond accuracy. Our study provides evidence that compression techniques significantly improve the efficiency and effectiveness of small-data language models, confirming that the prevailing beliefs regarding the effects of compression on large, heavily parameterized models hold true for less-parameterized, small-data models.
title What Happens When Small Is Made Smaller? Exploring the Impact of Compression on Small Data Pretrained Language Models
topic Computation and Language
Machine Learning
url https://arxiv.org/abs/2404.04759