Task-Driven Causal Feature Distillation: Towards Trustworthy Risk Prediction
This addresses the need for trustworthy and interpretable risk prediction in AI applications, but appears incremental as it builds on existing causal and distillation methods.
The paper tackled the problem of poor precision and recall in risk prediction models due to lack of causal reasoning and class imbalance, proposing a Task-Driven Causal Feature Distillation model that transforms features into causal attributions and uses a deep neural network, resulting in demonstrated superiority over state-of-the-art methods in precision, recall, interpretability, and causality.
Since artificial intelligence has seen tremendous recent successes in many areas, it has sparked great interest in its potential for trustworthy and interpretable risk prediction. However, most models lack causal reasoning and struggle with class imbalance, leading to poor precision and recall. To address this, we propose a Task-Driven Causal Feature Distillation model (TDCFD) to transform original feature values into causal feature attributions for the specific risk prediction task. The causal feature attribution helps describe how much contribution the value of this feature can make to the risk prediction result. After the causal feature distillation, a deep neural network is applied to produce trustworthy prediction results with causal interpretability and high precision/recall. We evaluate the performance of our TDCFD method on several synthetic and real datasets, and the results demonstrate its superiority over the state-of-the-art methods regarding precision, recall, interpretability, and causality.