Title: The Shapley Value of Phylogenetic Trees
Title: The Shapley Value of Phylogenetic Trees
Class: Distribution AI
Date: 10/2//23
The main hypothesis of the paper is that the Shapley value, a solution concept for
cooperative games, can be applied to phylogenetic trees to quantify the average marginal
diversity that each species brings to a group. The authors suggest that the Shapley value can
efforts..(Hypothesis, result)
The paper provides a mathematical analysis of the Shapley value on tree games, where the
players are represented by the leaves of a phylogenetic tree and the value of a coalition is the
total weight of the subtree spanned by the members of the coalition. The authors determine
the linear transformation that shows how the Shapley value depends on the edge weights of
the tree and compute a null space basis that allows for varying the edge weights without
changing the Shapley value. They also characterize the Shapley value on tree games by four
games.matrices.(summary,distribution)
I agree with the premise of the paper that the Shapley value can be applied to phylogenetic
trees to measure the contribution of each species to the diversity of a group. The authors
provide a clear explanation of the concept and its potential application in the field of
biodiversity preservation. However, I have a problem with the paper's thesis that the Shapley
value is the most important solution concept for n-player cooperative games. While the
Shapley value is certainly an important and widely used solution concept, there are other
solution concepts that are also widely used and have their own merits. For example, the
Banzhaf power index and the nucleolus are also commonly used solution concepts for
cooperative games.Furthermore, the Shapley value is the unique imputation vector that
satisfies four "fairness" criteria. While it is true that the Shapley value satisfies these criteria,
it is not the only solution concept that does so. For example, the nucleolus also satisfies these
criteria. Additionally, the paper could benefit from a discussion of the computational
complexity of calculating the Shapley value for tree games. The authors mention that they
compute the Shapley value on small trees, but it would be interesting to know if there are any
known algorithms or heuristics for efficiently computing the Shapley value on larger trees.
From this paper, I learned about the concept of the Shapley value in cooperative game
theory and its application to phylogenetic trees. The Shapley value is a solution concept that
assigns a value to each player in a cooperative game, based on their contributions to different
coalitions. In the context of phylogenetic trees, where the players are species and the
coalitions represent subsets of species, the Shapley value can be used to measure the average
marginal diversity that a species brings to any group. Furthermore, the paper suggests a
biological interpretation for the Shapley value in the context of biodiversity preservation. It
proposes that the Shapley value can be used as a ranking criterion to prioritize species for
preservation, based on their contributions to the overall diversity of a group. Overall, this
paper provides interesting insights into the application of game theory concepts to the
analysis of phylogenetic trees and sheds light on the potential use of the Shapley value in
biodiversity research.
Citation: Haake, C-J., Kashiwada, A., & Su, F. E. (2008). Mathematical biology. Journal
of Mathematical Biology, 56(3), 479-497. DOI: 10.1007/s00285-007-0126-2.