How optimal transport can tackle gender biases in multi-class neural-network classifiers for job recommendations?
This addresses bias mitigation in high-risk AI systems like job recommendation, which is important for fairness in employment but is incremental as it builds on existing optimal transport methods.
The authors tackled gender bias in multi-class neural-network classifiers for job recommendations by proposing a novel optimal transport strategy, which reduced algorithmic biases to lower levels than a standard method on the Bios dataset.
Automatic recommendation systems based on deep neural networks have become extremely popular during the last decade. Some of these systems can however be used for applications which are ranked as High Risk by the European Commission in the A.I. act, as for instance for online job candidate recommendation. When used in the European Union, commercial AI systems for this purpose will then be required to have to proper statistical properties with regard to potential discrimination they could engender. This motivated our contribution, where we present a novel optimal transport strategy to mitigate undesirable algorithmic biases in multi-class neural-network classification. Our stratey is model agnostic and can be used on any multi-class classification neural-network model. To anticipate the certification of recommendation systems using textual data, we then used it on the Bios dataset, for which the learning task consists in predicting the occupation of female and male individuals, based on their LinkedIn biography. Results show that it can reduce undesired algorithmic biases in this context to lower levels than a standard strategy.