Model independent feature attributions: Shapley values that uncover non-linear dependencies
نویسندگان
چکیده
منابع مشابه
Weighted Shapley levels values
This paper presents a collection of four different classes of weighted Shapley levels values. All classes contain generalisations of the weighted Shapley values to cooperative games with a level structure. The first class is an upgrade of the weighted Shapley levels value in Gómez-Rúa and Vidal-Puga (2011), who use the size of components as weights. The following classes contain payoff vectors ...
متن کاملShapley Inconsistency Values
There are relatively few proposals for inconsistency measures for propositional belief bases. However inconsistency measures are potentially as important as information measures for artificial intelligence, and more generally for computer science. In particular, they can be useful to define various operators for belief revision, belief merging, and negotiation. The measures that have been propo...
متن کاملFirst step immersion in interval linear programming with linear dependencies
We consider a linear programming problem in a general form and suppose that all coefficients may vary in some prescribed intervals. Contrary to classical models, where parameters can attain any value from the interval domains independently, we study problems with linear dependencies between the parameters. We present a class of problems that are easily solved by reduction to the classi...
متن کاملOn Modelling Non-linear Topical Dependencies
Probabilistic topic models such as Latent Dirichlet Allocation (LDA) discover latent topics from large corpora by exploiting words’ co-occurring relation. By observing the topical similarity between words, we find that some other relations, such as semantic or syntax relation between words, lead to strong dependence between their topics. In this paper, sentences are represented as dependency tr...
متن کاملWhen values and attributions collide: liberals' and conservatives' values motivate attributions for alleged misdeeds.
Conservatives tend to make dispositional whereas liberals make situational attributions for social problems and alleged misconduct (the "ideo-attribution effect"). Three studies demonstrated a reversal of the ideo-attribution effect. Conservatives made stronger situational attributions than liberals for the behavior of Marines accused of killing Iraqi civilians (Studies 1 and 2) and police offi...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: PeerJ Computer Science
سال: 2021
ISSN: 2376-5992
DOI: 10.7717/peerj-cs.582