Conformal Risk Training: End-to-End Optimization of Conformal Risk Control
This work addresses the need for reliable and risk-controlled AI deployments in domains such as healthcare and finance, offering a novel end-to-end optimization method rather than an incremental improvement.
The paper tackles the problem of providing provable risk guarantees for deep learning models in high-stakes applications by developing a method to control Optimized Certainty-Equivalent (OCE) risks, including tail risks like CVaR, and introduces an end-to-end training approach that improves average-case performance over post-hoc methods.
While deep learning models often achieve high predictive accuracy, their predictions typically do not come with any provable guarantees on risk or reliability, which are critical for deployment in high-stakes applications. The framework of conformal risk control (CRC) provides a distribution-free, finite-sample method for controlling the expected value of any bounded monotone loss function and can be conveniently applied post-hoc to any pre-trained deep learning model. However, many real-world applications are sensitive to tail risks, as opposed to just expected loss. In this work, we develop a method for controlling the general class of Optimized Certainty-Equivalent (OCE) risks, a broad class of risk measures which includes as special cases the expected loss (generalizing the original CRC method) and common tail risks like the conditional value-at-risk (CVaR). Furthermore, standard post-hoc CRC can degrade average-case performance due to its lack of feedback to the model. To address this, we introduce "conformal risk training," an end-to-end approach that differentiates through conformal OCE risk control during model training or fine-tuning. Our method achieves provable risk guarantees while demonstrating significantly improved average-case performance over post-hoc approaches on applications to controlling classifiers' false negative rate and controlling financial risk in battery storage operation.