A Value for n-person Games

By Lloyd S. Shapley. 1988

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Game & Decision Theory

Citations in the corpus

Explaining Models by Propagating Shapley Values of Local Components (2019, methodology): “In this paper, we focus on SHAP values (Lundberg and Lee 2017) – Shapley values (Shapley 1953) with a conditional expectation of the model prediction as the set function.”
Visualizing the Feature Importance for Black Box Models (2018, methodology): “y a set of players P, which can form coalitions SP. Each coalition Sachieves a certain payout. The characteristic function v: 2P !R maps all 2ppossible coalitions to their payouts. The Shapley value [19] now fairly assigns a value to each player depending on their contribution in all possible coalit”
A Unified Approach to Interpreting Model Predictions (2017, background): “Theorem 1 follows from a combination of results in cooperative game theory, where the values φ i are known as the Shapley values [5].”
Explainable machine learning in deployment (2019, methodology): “offeature-level explanationthatdenotehowimportant a region or collection of features, is [5, 23]. A prominent class of perturbation based methods are based on Shapley values from cooperativegametheory[50].Shapleyvaluesareafairwaytodistribute the gains from a cooperative game to its players. In apply”
Opportunities and obstacles for deep learning in biology and medicine (2017, background): “Lundberg and Lee [342] noted that several importance scoring methods including integrated gradients and LIME could all be considered approximations to Shapely values [343], which have a long history in game theory for assigning contributions to players in cooperative games.”
Online Class-Incremental Continual Learning with Adversarial Shapley Value (2020, methodology): “The SV (Shapley 1953; Roth 1988) was originally proposed in cooperative game theory to decide the share of total gains for each player in a coalition.”
Data Shapley: Equitable Valuation of Data for Machine Learning (2019, methodology): “Shapley value was proposed in a classic paper in game theory (Shapley, 1953) and has been widely influential in economics (Shapley et al.”
Towards Efficient Data Valuation Based on the Shapley Value (2019, background): “The SV [26] is a classic concept in cooperative game theory to attribute the total gains generated by the coalition of all players.”
Understanding Global Feature Contributions With Additive Importance Measures (2020, methodology): “Research in game theory has extensively analyzed credit allocation for cooperative games, so we apply a game theoretic solution known as the Shapley value [33].”
Cooperative game approaches to coordinating a three-echelon closed-loop supply chain with fairness concerns (2019, background): “Shapley value (Shapley, 1953) is a widely accepted profit allocation mechanism in cooperative games, which is simply the average marginal contribution of each player if this player enters all possible coalitions in a completely random order.”
An Axiomatic Theory of Fairness in Network Resource Allocation (2009, background): “While applications of these theories range from political theory to voting methods, here we focus on two well-known axiomatic constructions: the the Nash bargaining solution [20] and the Shapley value [21].”
TOWARDS HIERARCHICAL IMPORTANCE ATTRIBU- (2019, methodology): “(b) Additive attributions assign importance scores for words “not” and “interesting” by linearly distributing contribution score of “not interesting“, exemplified with Shapley Values (Shapley, 1997).”
Is Investment-Cash Flow Sensitivity Caused by Agency Costs or Asymmetric Information? Evidence from the UK (2005, unknown): “Therefore, we resort to Shapley (1953) values (SVs), which assign a power index to each shareholder that reflects their relative importance in forming winning voting coalitions.”
A Survey of Deep Learning for Scientific Discovery (2020, background): “Arguably the most well known of the ablation based approaches is the notion of a Shapely value , ﬁrst introduced in [207].”
A Social Capital Index (2011, background): “A game v ∈ G N is a veto-rich game if it has at least one veto-player.”
Generalized Integrated Gradients: A practical method for explaining diverse ensembles (2019, methodology): “We use the terminology from Shapley (1953) and Aumann and Shapley (1974) where appropriate.”