ARDA: automatic relational data augmentation for machine learning
Name
2003.09758.pdf
Description
Accepted version
Size
609.21 KB
Format
Unknown
Checksum (MD5)
c6cd474ea4d2e2e4c8d0c39952678656
Author(s)
Chepurko, Nadiia
Marcus, Ryan
Zgraggen, Emanuel
Castro Fernandez, Raul
Kraska, Tim
Karger, David R
Date Issued
2020
Journal
Proceedings of the VLDB Endowment
Publisher
VLDB Endowment
Version
Original manuscript
Abstract
© 2020, VLDB Endowment. Automatic machine learning (AML) is a family of techniques to automate the process of training predictive models, aim-ing to both improve performance and make machine learn-ing more accessible. While many recent works have focused on aspects of the machine learning pipeline like model se-lection, hyperparameter tuning, and feature selection, rela-tively few works have focused on automatic data augmen-tation. Automatic data augmentation involves finding new features relevant to the user's predictive task with minimal "human-in-the-loop" involvement. We present ARDA, an end-to-end system that takes as input a dataset and a data repository, and outputs an aug-mented data set such that training a predictive model on this augmented dataset results in improved performance. Our system has two distinct components: (1) a framework to search and join data with the input data, based on various attributes of the input, and (2) an efficient feature selection algorithm that prunes out noisy or irrelevant features from the resulting join. We perform an extensive empirical eval-uation of different system components and benchmark our feature selection algorithm on real-world datasets.
MIT Department
Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory
Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science
Terms of Use
Creative Commons Attribution-Noncommercial-Share Alike
Persistent DSpace Link
DOI of Published Version
https://doi.org/10.14778/3397230.3397235
