Understanding Black-box Predictions via Influence Functions

2017 inproceedings koh2017understanding ○ Not yet verified

Authors

Pang Wei Koh, Percy Liang

Venue

Proceedings of the 34th International Conference on Machine Learning (ICML)

Abstract

Uses influence functions from robust statistics to trace model predictions back to training data, identifying training points most responsible for a given prediction.

BibTeX

Local Entry

@inproceedings{koh2017understanding,
  title = {Understanding Black-box Predictions via Influence Functions},
  author = {Pang Wei Koh and Percy Liang},
  year = {2017},
  booktitle = {Proceedings of the 34th International Conference on Machine Learning (ICML)},
  url = {https://proceedings.mlr.press/v70/koh17a.html},
  abstract = {Uses influence functions from robust statistics to trace model predictions back to training data, identifying training points most responsible for a given prediction.}
}

From OPENALEX

@inproceedings{koh2017understanding,
  title = {Understanding Black-box Predictions via Influence Functions},
  author = {Pang Wei Koh and Percy Liang},
  year = {2017},
  booktitle = {arXiv (Cornell University)},
  doi = {10.48550/arxiv.1703.04730}
}