Priority-Aware Shapley Value
This addresses a limitation in model-agnostic data valuation and feature attribution for ML practitioners, offering an incremental improvement by extending Shapley values with priority awareness.
The paper tackled the problem that Shapley values assume contributors are interchangeable, which can be problematic for dependent data or when contributions need adjustment by factors like trust or risk, by proposing Priority-Aware Shapley Value (PASV) that incorporates precedence constraints and priority weights, resulting in more structure-faithful allocations in experiments on MNIST/CIFAR10 and Census Income datasets.
Shapley values are widely used for model-agnostic data valuation and feature attribution, yet they implicitly assume contributors are interchangeable. This can be problematic when contributors are dependent (e.g., reused/augmented data or causal feature orderings) or when contributions should be adjusted by factors such as trust or risk. We propose Priority-Aware Shapley Value (PASV), which incorporates both hard precedence constraints and soft, contributor-specific priority weights. PASV is applicable to general precedence structures, recovers precedence-only and weight-only Shapley variants as special cases, and is uniquely characterized by natural axioms. We develop an efficient adjacent-swap Metropolis-Hastings sampler for scalable Monte Carlo estimation and analyze limiting regimes induced by extreme priority weights. Experiments on data valuation (MNIST/CIFAR10) and feature attribution (Census Income) demonstrate more structure-faithful allocations and a practical sensitivity analysis via our proposed "priority sweeping".