Revisiting Explicit Regularization in Neural Networks for Well-Calibrated Predictive Uncertainty
This addresses the need for reliable uncertainty estimates in deterministic neural networks, offering a scalable solution for applications requiring calibrated predictions, though it appears incremental by building on existing regularization concepts.
The paper tackles the problem of improving predictive uncertainty calibration in neural networks by revisiting explicit regularization, showing that it can enhance log-likelihood and provide well-calibrated uncertainty as an efficient alternative to Bayesian methods.
From the statistical learning perspective, complexity control via explicit regularization is a necessity for improving the generalization of over-parameterized models. However, the impressive generalization performance of neural networks with only implicit regularization may be at odds with this conventional wisdom. In this work, we revisit the importance of explicit regularization for obtaining well-calibrated predictive uncertainty. Specifically, we introduce a probabilistic measure of calibration performance, which is lower bounded by the log-likelihood. We then explore explicit regularization techniques for improving the log-likelihood on unseen samples, which provides well-calibrated predictive uncertainty. Our findings present a new direction to improve the predictive probability quality of deterministic neural networks, which can be an efficient and scalable alternative to Bayesian neural networks and ensemble methods.