Comment on page
Option 1: Pre-processing
If we decide to mitigate bias at the pre-processing stage, we modify the training data in order to reduce or remove unwanted bias before training the model (e.g. re-sampling). The aim is to train the model on data that is free from unwanted bias. Pre-processing methods are agnostic, which means that the type of model does not matter.
- Reweighting, where the tuples in the training set are assigned weights. By modulating the weights, and using metrics based on frequency counts, it is possible to remove discrimination from the dataset (Kamiran and Calders, 2012)
- Learning data representations, where the aim of the model is to get to a representation of the data that both encodes the data as well as possible, and also obfuscates any information about membership in the protected group. (Zemel et al 2013)