Score-Based Explanations in Data Management and Machine Learning
Databases
2020-08-20 v2 Artificial Intelligence
Machine Learning
Abstract
We describe some approaches to explanations for observed outcomes in data management and machine learning. They are based on the assignment of numerical scores to predefined and potentially relevant inputs. More specifically, we consider explanations for query answers in databases, and for results from classification models. The described approaches are mostly of a causal and counterfactual nature. We argue for the need to bring domain and semantic knowledge into score computations; and suggest some ways to do this.
Cite
@article{arxiv.2007.12799,
title = {Score-Based Explanations in Data Management and Machine Learning},
author = {Leopoldo Bertossi},
journal= {arXiv preprint arXiv:2007.12799},
year = {2020}
}
Comments
Companion paper for a tutorial at the Scalable Uncertainty Management Conference (SUM'20). To appear in Proc. SUM'20. Minor fixes made