🤖 AI Summary
Existing Shapley-based data valuation methods neglect the intrinsic distributional characteristics of data value and fail to model dynamic data conditions, leading to significant estimation bias and low computational efficiency. To address these limitations, we propose a distribution-aware Shapley valuation framework: (i) we systematically model both global and local statistical distributions of data value for the first time; (ii) building upon this, we enhance Additive Margin Estimation (AME) to develop a dynamic valuation paradigm adaptable to data addition/removal, label noise, and incremental/decremental learning scenarios. Our approach unifies Shapley theory, distributional modeling, and optimization-based inference. Empirical results demonstrate that it substantially outperforms state-of-the-art methods in Shapley value estimation accuracy, label error detection, and evaluation of data perturbations—achieving both high fidelity and computational efficiency.
📝 Abstract
Data valuation has garnered increasing attention in recent years, given the critical role of high-quality data in various applications, particularly in machine learning tasks. There are diverse technical avenues to quantify the value of data within a corpus. While Shapley value-based methods are among the most widely used techniques in the literature due to their solid theoretical foundation, the accurate calculation of Shapley values is often intractable, leading to the proposal of numerous approximated calculation methods. Despite significant progress, nearly all existing methods overlook the utilization of distribution information of values within a data corpus. In this paper, we demonstrate that both global and local statistical information of value distributions hold significant potential for data valuation within the context of machine learning. Firstly, we explore the characteristics of both global and local value distributions across several simulated and real data corpora. Useful observations and clues are obtained. Secondly, we propose a new data valuation method that estimates Shapley values by incorporating the explored distribution characteristics into an existing method, AME. Thirdly, we present a new path to address the dynamic data valuation problem by formulating an optimization problem that integrates information of both global and local value distributions. Extensive experiments are conducted on Shapley value estimation, value-based data removal/adding, mislabeled data detection, and incremental/decremental data valuation. The results showcase the effectiveness and efficiency of our proposed methodologies, affirming the significant potential of global and local value distributions in data valuation.