A Model Explanation System: Latest Updates and Extensions

By Ryan D. Turner. 2016

15 citations recorded by the source corpus.

We propose a general model explanation system (MES) for "explaining" the output of black box classifiers. This paper describes extensions to Turner (2015), which is referred to frequently in the text. We use the motivating example of a classifier trained to detect fraud in a credit card transaction history. The key aspect is that we provide explanations applicable to a single prediction, rather than provide an interpretable set of parameters. We focus on explaining positive predictions (alerts). However, the presented methodology is symmetrically applicable to negative predictions.

Subfields: Artificial Intelligence, Management Science and Operations Research

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

In Management Science and Operations Research

Citations in the corpus

Increasing the Interpretability of Recurrent Neural Networks Using Hidden Markov Models (2016, background): “n be considered intelligible to humans in multiple ways, falling under the broad categories of transparency and post-hoc interpretability (Lipton, 2016). While works such as Ribeiro et al. (2016) and Turner (2016) develop post-hoc explanations for black-box models, we focus on transparency, speciﬁca”
Principles and Practice of Explainable Machine Learning (2020, background): “Additional techniques to construct rules explaining a model’s decisions can be found in [82, 83].”
Counterfactual Explanations and Algorithmic Recourses for Machine Learning: A Review (2020, unknown): “Model-specific approaches for deep neural networks (CAM, Grad-CAM [274, 355]), and model agnostic approaches (LIME, MES [261, 307]) have been”
Counterfactual Explanations for Machine Learning: A Review (2020, methodology): “ion need not be a global explanation or explain the internal logic of the model. Model-specific approaches for deep neural networks (CAM, Grad-CAM [96, 115]), and model agnostic approaches (LIME, MES [92, 106]) have been proposed. These are either feature attribution or model simplification methods.”