Uncertainty Baselines: Benchmarks for Uncertainty & Robustness in Deep Learning
This provides standardized benchmarks for researchers and practitioners to evaluate and compare uncertainty and robustness techniques in deep learning, though it is incremental as it consolidates existing methods rather than proposing new ones.
The paper introduces Uncertainty Baselines, a collection of high-quality implementations for 19 methods across 9 tasks to benchmark uncertainty and robustness in deep learning, addressing gaps in competitive comparisons due to compute, baselines, and reproducibility issues.
High-quality estimates of uncertainty and robustness are crucial for numerous real-world applications, especially for deep learning which underlies many deployed ML systems. The ability to compare techniques for improving these estimates is therefore very important for research and practice alike. Yet, competitive comparisons of methods are often lacking due to a range of reasons, including: compute availability for extensive tuning, incorporation of sufficiently many baselines, and concrete documentation for reproducibility. In this paper we introduce Uncertainty Baselines: high-quality implementations of standard and state-of-the-art deep learning methods on a variety of tasks. As of this writing, the collection spans 19 methods across 9 tasks, each with at least 5 metrics. Each baseline is a self-contained experiment pipeline with easily reusable and extendable components. Our goal is to provide immediate starting points for experimentation with new methods or applications. Additionally we provide model checkpoints, experiment outputs as Python notebooks, and leaderboards for comparing results. Code available at https://github.com/google/uncertainty-baselines.