Robust Data Sampling in Machine Learning: A Game-Theoretic Framework for Training and Validation Data Selection

Zhaobin Mo, Xuan Di () and Rongye Shi
Additional contact information
Zhaobin Mo: Department of Civil Engineering and Engineering Mechanics, Columbia University, New York, NY 10027, USA
Xuan Di: Department of Civil Engineering and Engineering Mechanics, Columbia University, New York, NY 10027, USA
Rongye Shi: Department of Civil Engineering and Engineering Mechanics, Columbia University, New York, NY 10027, USA

Games, 2023, vol. 14, issue 1, 1-13

Abstract: How to sample training/validation data is an important question for machine learning models, especially when the dataset is heterogeneous and skewed. In this paper, we propose a data sampling method that robustly selects training/validation data. We formulate the training/validation data sampling process as a two-player game: a trainer aims to sample training data so as to minimize the test error, while a validator adversarially samples validation data that can increase the test error. Robust sampling is achieved at the game equilibrium. To accelerate the searching process, we adopt reinforcement learning aided Monte Carlo trees search (MCTS). We apply our method to a car-following modeling problem, a complicated scenario with heterogeneous and random human driving behavior. Real-world data, the Next Generation SIMulation (NGSIM), is used to validate this method, and experiment results demonstrate the sampling robustness and thereby the model out-of-sample performance.

Keywords: two-player game; Monte Carlo tree search; reinforcement learning; car-following modeling (search for similar items in EconPapers)
JEL-codes: C C7 C70 C71 C72 C73 (search for similar items in EconPapers)
Date: 2023
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
https://www.mdpi.com/2073-4336/14/1/13/pdf (application/pdf)
https://www.mdpi.com/2073-4336/14/1/13/ (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:gam:jgames:v:14:y:2023:i:1:p:13-:d:1051349

Access Statistics for this article

Games is currently edited by Ms. Susie Huang

More articles in Games from MDPI
Bibliographic data for series maintained by MDPI Indexing Manager ().