- Tags:: 📜Papers
- Authors: Janis Klaise, Arnaud Van Looveren, Clive Cox, Giovanni Vacanti, Alexandru Coca
- Link:: http://arxiv.org/abs/2007.06299
- Zotero Link:: klaiseMonitoringExplainabilityModels2020
- Source date:: 2020
- Finished date:: 2021-07-13
There are three areas of interest:
- The usual Model Performance, for example latency.
- Model Outlier Detection: that is, for a single instance, whether it is out-of-distribution of training data.
- And Model Drifting, which can be further distinguished in two:
- Covariate Shift: the input data distribution changes, but p(y|x) remains the same.
- Label Shift: the label distribution changes (that is, reality) but the conditional distribution stays the same p(x|y)