A multi-task network approach for calculating discrimination-free insurance prices
This addresses the issue of proxy discrimination in insurance pricing for insurers and regulators, offering a technical solution that is incremental by building on existing methods to handle missing protected data.
The paper tackled the problem of indirect discrimination in insurance pricing by developing a multi-task neural network that can be trained with only partial information on protected characteristics, achieving predictive accuracy comparable to a conventional neural network on full data and superior performance when information is partially missing.
In applications of predictive modeling, such as insurance pricing, indirect or proxy discrimination is an issue of major concern. Namely, there exists the possibility that protected policyholder characteristics are implicitly inferred from non-protected ones by predictive models, and are thus having an undesirable (or illegal) impact on prices. A technical solution to this problem relies on building a best-estimate model using all policyholder characteristics (including protected ones) and then averaging out the protected characteristics for calculating individual prices. However, such approaches require full knowledge of policyholders' protected characteristics, which may in itself be problematic. Here, we address this issue by using a multi-task neural network architecture for claim predictions, which can be trained using only partial information on protected characteristics, and it produces prices that are free from proxy discrimination. We demonstrate the use of the proposed model and we find that its predictive accuracy is comparable to a conventional feedforward neural network (on full information). However, this multi-task network has clearly superior performance in the case of partially missing policyholder information.