Socrates Loss: Unifying Confidence Calibration and Classification by Leveraging the Unknown
This work addresses the reliability of deep neural networks in high-stakes applications by improving confidence calibration without sacrificing classification performance.
The paper introduces Socrates Loss, a unified loss function that leverages an auxiliary unknown class to simultaneously optimize deep neural networks for classification accuracy and confidence calibration, mitigating the stability-performance trade-off. Experiments across four datasets show consistent improvements in accuracy-calibration trade-offs and faster convergence.
Deep neural networks, despite their high accuracy, often exhibit poor confidence calibration, limiting their reliability in high-stakes applications. Current ad-hoc confidence calibration methods attempt to fix this during training but face a fundamental trade-off: two-phase training methods achieve strong classification performance at the cost of training instability and poorer confidence calibration, while single-loss methods are stable but underperform in classification. This paper addresses and mitigates this stability-performance trade-off. We propose Socrates Loss, a novel, unified loss function that explicitly leverages uncertainty by incorporating an auxiliary unknown class, whose predictions directly influence the loss function and a dynamic uncertainty penalty. This unified objective allows the model to be optimized for both classification and confidence calibration simultaneously, without the instability of complex, scheduled losses. We provide theoretical guarantees that our method regularizes the model to prevent miscalibration and overfitting. Across four benchmark datasets and multiple architectures, our comprehensive experiments demonstrate that Socrates Loss consistently improves training stability while achieving more favorable accuracy-calibration trade-off, often converging faster than existing methods.