Efficient Index Learning via Model Reuse and Fine-tuning

被引：0

作者：

Liu, Guanli ^{[1
]}

Qi, Jianzhong ^{[1
]}

Kulik, Lars ^{[1
]}

Soga, Kazuya ^{[1
]}

Borovica-Gajic, Renata ^{[1
]}

Rubinstein, Benjamin I. P. ^{[1
]}

机构：

[1] Univ Melbourne, Melbourne, Vic, Australia

来源：

2023 IEEE 39TH INTERNATIONAL CONFERENCE ON DATA ENGINEERING WORKSHOPS, ICDEW | 2023年

基金：

澳大利亚研究理事会;

关键词：

learned index; model reuse; fine-tuning;

D O I：

10.1109/ICDEW58674.2023.00015

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Learned indices using machine learning techniques have demonstrated potential as alternatives to traditional indices such as B-trees in both query time and memory. However, a well fitted learned index requires significant space consumption to train models and tune parameters. Furthermore, fast training methods-ones that train in one pass-may not learn the data distribution well. To consider both the fitness to data distribution and building efficiency, in this paper, we apply pre-trained models and fine-tuning to accelerate the building of learned indices by 30.4% and improve lookup efficiency by up to 24.4% on real datasets and 22.5% on skewed datasets.

引用

页码：60 / 66

页数：7

共 50 条

[1] How fine can fine-tuning be? Learning efficient language models
Radiya-Dixit, Evani
Wang, Xin
[J]. INTERNATIONAL CONFERENCE ON ARTIFICIAL INTELLIGENCE AND STATISTICS, VOL 108, 2020, 108 : 2435 - 2442
[2] An efficient deep learning model to categorize brain tumor using reconstruction and fine-tuning
Talukder, Md. Alamin
Islam, Md. Manowarul
Uddin, Md. Ashraf
Akhter, Arnisha
Pramanik, Md. Alamgir Jalil
Aryal, Sunil
Almoyad, Muhammad Ali Abdulllah
Hasan, Khondokar Fida
Moni, Mohammad Ali
[J]. EXPERT SYSTEMS WITH APPLICATIONS, 2023, 230
[3] Scalable Online Planning via Reinforcement Learning Fine-Tuning
Fickinger, Arnaud
Hu, Hengyuan
Amos, Brandon
Russell, Stuart
Brown, Noam
[J]. ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 34 (NEURIPS 2021), 2021, 34
[4] Efficient Unsupervised Sentence Compression by Fine-tuning Transformers with Reinforcement Learning
Ghalandari, Demian Gholipour
Hokamp, Chris
Ifrim, Georgiana
[J]. PROCEEDINGS OF THE 60TH ANNUAL MEETING OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS (ACL 2022), VOL 1: (LONG PAPERS), 2022, : 1267 - 1280
[5] Transfer Learning With Adaptive Fine-Tuning
Vrbancic, Grega
Podgorelec, Vili
[J]. IEEE ACCESS, 2020, 8 : 196197 - 196211
[6] Fine-Tuning via Mask Language Model Enhanced Representations Based Contrastive Learning and Application
Zhang, Dechi
Wan, Weibing
[J]. Computer Engineering and Applications, 2024, 60 (17) : 129 - 138
[7] Towards Adaptive Prefix Tuning for Parameter-Efficient Language Model Fine-tuning
Zhang, Zhen-Ru
Tan, Chuanqi
Xu, Haiyang
Wang, Chengyu
Huang, Jun
Huang, Songfang
[J]. 61ST CONFERENCE OF THE THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS, ACL 2023, VOL 2, 2023, : 1239 - 1248
[8] MultiFiT: Efficient Multi-lingual Language Model Fine-tuning
Eisenschlos, Julian
Ruder, Sebastian
Czapla, Piotr
Kardas, Marcin
Gugger, Sylvain
Howard, Jeremy
[J]. 2019 CONFERENCE ON EMPIRICAL METHODS IN NATURAL LANGUAGE PROCESSING AND THE 9TH INTERNATIONAL JOINT CONFERENCE ON NATURAL LANGUAGE PROCESSING (EMNLP-IJCNLP 2019): PROCEEDINGS OF THE CONFERENCE, 2019, : 5702 - 5707
[9] PockEngine: Sparse and Efficient Fine-tuning in a Pocket
Zhu, Ligeng
Hu, Lanxiang
Lin, Ji
Wang, Wei-Chen
Chen, Wei-Ming
Gan, Chuang
Han, Song
[J]. 56TH IEEE/ACM INTERNATIONAL SYMPOSIUM ON MICROARCHITECTURE, MICRO 2023, 2023, : 1381 - 1394
[10] Efficient Fine-Tuning of BERT Models on the Edge
Vucetic, Danilo
Tayaranian, Mohammadreza
Ziaeefard, Maryam
Clark, James J.
Meyer, Brett H.
Gross, Warren J.
[J]. 2022 IEEE INTERNATIONAL SYMPOSIUM ON CIRCUITS AND SYSTEMS (ISCAS 22), 2022, : 1838 - 1842

← 1 2 3 4 5 →