Calibrated Reliable Regression using Maximum Mean Discrepancy
This work addresses the need for accurate uncertainty quantification in machine learning applications, offering a domain-specific improvement for regression tasks.
The paper tackled the problem of unreliable predictive uncertainty in deep neural networks for regression tasks, proposing a calibrated regression method using maximum mean discrepancy that asymptotically converges to zero calibration error and outperforms state-of-the-art methods on real datasets.
Accurate quantification of uncertainty is crucial for real-world applications of machine learning. However, modern deep neural networks still produce unreliable predictive uncertainty, often yielding over-confident predictions. In this paper, we are concerned with getting well-calibrated predictions in regression tasks. We propose the calibrated regression method using the maximum mean discrepancy by minimizing the kernel embedding measure. Theoretically, the calibration error of our method asymptotically converges to zero when the sample size is large enough. Experiments on non-trivial real datasets show that our method can produce well-calibrated and sharp prediction intervals, which outperforms the related state-of-the-art methods.