Towards Aggregating Weighted Feature Attributions

By Umang Bhatt, Pradeep Ravikumar, José M. F. Moura. 2019

14 citations recorded by the source corpus.

Current approaches for explaining machine learning models fall into two distinct classes: antecedent event influence and value attribution. The former leverages training instances to describe how much influence a training point exerts on a test point, while the latter attempts to attribute value to the features most pertinent to a given prediction. In this work, we discuss an algorithm, AVA: Aggregate Valuation of Antecedents, that fuses these two explanation classes to form a new approach to feature attribution that not only retrieves local explanations but also captures global patterns learned by a model. Our experimentation convincingly favors weighting and aggregating feature attributions via AVA.

Subfields: Artificial Intelligence

How researchers see this paper

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

In Artificial Intelligence

Citations in the corpus

Explainable machine learning in deployment (2019, methodology): “y value is exponential in d, input dimensionality; however, the literature has proposed approximate methods using weighted linear regression [34], Monte Carlo approximation [61], centroid aggregation [11], and graph-structured factorization [14]. When we refer to Shapley-related methods hereafter, w”
Evaluating and Aggregating Feature-based Model Explanations (2020, methodology): “This is a natural analog for how humans reason: we use past similar events (training data) and facts about the present (individual features) to make decisions [Bhatt et al. , 2019b].”
A Survey on the Explainability of Supervised Machine Learning (2020, background): “Aggregate Valuation of Antecedents (AVA) (Bhatt et al., 2019) is a value attribution technique that provides local explanations but also detects global patterns.”