• Corpus ID: 233204739

Individual Explanations in Machine Learning Models: A Survey for Practitioners

@article{Carrillo2021IndividualEI,
  title={Individual Explanations in Machine Learning Models: A Survey for Practitioners},
  author={Alfredo Carrillo and Luis F. Cant'u and Alejandro Noriega},
  journal={ArXiv},
  year={2021},
  volume={abs/2104.04144}
}
In recent years, the use of sophisticated statistical models that influence decisions in domains of high societal relevance is on the rise. Although these models can often bring substantial improvements in the accuracy and efficiency of organizations, many governments, institutions, and companies are reluctant to their adoption as their output is often difficult to explain in human-interpretable ways. Hence, these models are often regarded as black-boxes, in the sense that their internal… 

Tables from this paper

Model Explanations via the Axiomatic Causal Lens

This work proposes three explanation measures which aggregate the set of all but-for causes — a necessary and sufficient explanation — into feature importance weights, and is the first to formally bridge the gap between model explanations, game-theoretic influence, and causal analysis.

Individual Explanations in Machine Learning Models: A Case Study on Poverty Estimation

This case study presents a set of strategies that mitigate challenges, as faced when implementing explanation methods in a relevant application domain—poverty estimation and its use for prioritizing access to social policies.

References

SHOWING 1-10 OF 50 REFERENCES

Interpretable Machine Learning

This project introduces Robust T CAV, which builds on TCAV and experimentally determines best practices for this method and is a step in the direction of making TCAVs, an already impactful algorithm in interpretability, more reliable and useful for practitioners.

Explainable machine-learning predictions for the prevention of hypoxaemia during surgery

The results suggest that if anaesthesiologists currently anticipate 15% of hypoxaemia events, with the assistance of this system they could anticipate 30%, a large portion of which may benefit from early intervention because they are associated with modifiable factors.

Real Time Image Saliency for Black Box Classifiers

A masking model is trained to manipulate the scores of the classifier by masking salient parts of the input image to generalise well to unseen images and requires a single forward pass to perform saliency detection, therefore suitable for use in real-time systems.

Interpretable Explanations of Black Boxes by Meaningful Perturbation

A general framework for learning different kinds of explanations for any black box algorithm is proposed and the framework to find the part of an image most responsible for a classifier decision is specialised.

Opening black box Data Mining models using Sensitivity Analysis

  • P. CortezM. Embrechts
  • Computer Science
    2011 IEEE Symposium on Computational Intelligence and Data Mining (CIDM)
  • 2011
This paper proposes a Global SA (GSA), which extends the applicability of previous SA methods, and several visualization techniques, for assessing input relevance and effects on the model's responses.

Explaining Classifications For Individual Instances

It is demonstrated that the generated explanations closely follow the learned models and a visualization technique is presented that shows the utility of the approach and enables the comparison of different prediction methods.

From local explanations to global understanding with explainable AI for trees

An explanation method for trees is presented that enables the computation of optimal local explanations for individual predictions, and the authors demonstrate their method on three medical datasets.

Contrastive Explanation

  • P. Lipton
  • Philosophy
    Royal Institute of Philosophy Supplement
  • 1990
According to a causal model of explanation, we explain phenomena by giving their causes or, where the phenomena are themselves causal regularities, we explain them by giving a mechanism linking cause

The Dangers of Post-hoc Interpretability: Unjustified Counterfactual Explanations

This paper evaluates the risk of generating unjustified counterfactual examples by investigating the local neighborhoods of instances whose predictions are to be explained and shows that this risk is quite high for several datasets.

Policing of Terrorism Using Data from Social Media

It is argued that technology-driven approaches do not fit with current practices of policing in the field of terrorism and extremism, which build on professional judgement rather than algorithm-driven pattern identification in big data.