Data Valuation using Reinforcement Learning

TL;DR

This paper introduces DVRL, a reinforcement learning-based framework for data valuation that adaptively estimates data importance to improve machine learning tasks such as domain adaptation and robust learning.

Contribution

The paper proposes a novel meta learning framework using reinforcement learning to estimate data value, outperforming existing methods across various applications.

Findings

DVRL provides superior data value estimates compared to alternatives.

Close to optimal corrupted sample discovery performance.

Significant improvements in domain adaptation and robust learning tasks.

Abstract

Quantifying the value of data is a fundamental problem in machine learning. Data valuation has multiple important use cases: (1) building insights about the learning task, (2) domain adaptation, (3) corrupted sample discovery, and (4) robust learning. To adaptively learn data values jointly with the target task predictor model, we propose a meta learning framework which we name Data Valuation using Reinforcement Learning (DVRL). We employ a data value estimator (modeled by a deep neural network) to learn how likely each datum is used in training of the predictor model. We train the data value estimator using a reinforcement signal of the reward obtained on a small validation set that reflects performance on the target task. We demonstrate that DVRL yields superior data value estimates compared to alternative methods across different types of datasets and in a diverse set of application scenarios. The corrupted sample discovery performance of DVRL is close to optimal in many regimes (i.e. as if the noisy samples were known apriori), and for domain adaptation and robust learning DVRL significantly outperforms state-of-the-art by 14.6% and 10.8%, respectively.