Beyond Deep Ensembles: A Large-Scale Evaluation of Bayesian Deep Learning under Distribution Shift
This work provides a systematic benchmark for researchers and practitioners in machine learning to compare Bayesian deep learning methods under realistic distribution shifts, though it is incremental as it extends existing evaluations.
The paper conducted a large-scale evaluation of Bayesian deep learning methods on real-world datasets from the WILDS collection to assess generalization and calibration under distribution shift, finding that ensembling improves performance but variational inference excels in fine-tuning large language models.
Bayesian deep learning (BDL) is a promising approach to achieve well-calibrated predictions on distribution-shifted data. Nevertheless, there exists no large-scale survey that evaluates recent SOTA methods on diverse, realistic, and challenging benchmark tasks in a systematic manner. To provide a clear picture of the current state of BDL research, we evaluate modern BDL algorithms on real-world datasets from the WILDS collection containing challenging classification and regression tasks, with a focus on generalization capability and calibration under distribution shift. We compare the algorithms on a wide range of large, convolutional and transformer-based neural network architectures. In particular, we investigate a signed version of the expected calibration error that reveals whether the methods are over- or under-confident, providing further insight into the behavior of the methods. Further, we provide the first systematic evaluation of BDL for fine-tuning large pre-trained models, where training from scratch is prohibitively expensive. Finally, given the recent success of Deep Ensembles, we extend popular single-mode posterior approximations to multiple modes by the use of ensembles. While we find that ensembling single-mode approximations generally improves the generalization capability and calibration of the models by a significant margin, we also identify a failure mode of ensembles when finetuning large transformer-based language models. In this setting, variational inference based approaches such as last-layer Bayes By Backprop outperform other methods in terms of accuracy by a large margin, while modern approximate inference algorithms such as SWAG achieve the best calibration.