Revisiting the Evaluation of Uncertainty Estimation and Its Application to Explore Model Complexity-Uncertainty Trade-Off
This work addresses the problem of improving uncertainty estimation for building trusted deep neural networks in applications like security and autonomous driving, but it is incremental as it builds on existing use cases and metrics.
The paper identifies issues with existing metrics for uncertainty estimation in selective prediction and confidence calibration, proposes new metrics to address them, and uses these to explore the trade-off between model complexity and uncertainty quality, with empirical results validating the new metrics and revealing trends.
Accurately estimating uncertainties in neural network predictions is of great importance in building trusted DNNs-based models, and there is an increasing interest in providing accurate uncertainty estimation on many tasks, such as security cameras and autonomous driving vehicles. In this paper, we focus on the two main use cases of uncertainty estimation, i.e. selective prediction and confidence calibration. We first reveal potential issues of commonly used quality metrics for uncertainty estimation in both use cases, and propose our new metrics to mitigate them. We then apply these new metrics to explore the trade-off between model complexity and uncertainty estimation quality, a critically missing work in the literature. Our empirical experiment results validate the superiority of the proposed metrics, and some interesting trends about the complexity-uncertainty trade-off are observed.