BLIND: Bias Removal With No Demographics
This addresses the challenge of mitigating social biases in AI models for applications like classification, reducing reliance on costly demographic annotations, though it is incremental as it builds on existing bias mitigation techniques.
The paper tackles the problem of social bias amplification in models trained on real-world data by introducing BLIND, a method that removes bias without requiring prior demographic knowledge, achieving competitive or superior performance to methods that need such information in sentiment and occupation classification tasks.
Models trained on real-world data tend to imitate and amplify social biases. Common methods to mitigate biases require prior information on the types of biases that should be mitigated (e.g., gender or racial bias) and the social groups associated with each data sample. In this work, we introduce BLIND, a method for bias removal with no prior knowledge of the demographics in the dataset. While training a model on a downstream task, BLIND detects biased samples using an auxiliary model that predicts the main model's success, and down-weights those samples during the training process. Experiments with racial and gender biases in sentiment classification and occupation classification tasks demonstrate that BLIND mitigates social biases without relying on a costly demographic annotation process. Our method is competitive with other methods that require demographic information and sometimes even surpasses them.