OoD-Bench: Quantifying and Understanding Two Dimensions of Out-of-Distribution Generalization
This work addresses the challenge of OoD generalization for machine learning practitioners by providing a systematic benchmark, though it is incremental in nature as it builds on existing datasets and algorithms.
The paper tackles the problem of out-of-distribution (OoD) generalization in deep learning by identifying and measuring two distinct types of distribution shifts in datasets, and it compares OoD algorithms across benchmarks to reveal their strengths and limitations, positioning existing research into a coherent framework.
Deep learning has achieved tremendous success with independent and identically distributed (i.i.d.) data. However, the performance of neural networks often degenerates drastically when encountering out-of-distribution (OoD) data, i.e., when training and test data are sampled from different distributions. While a plethora of algorithms have been proposed for OoD generalization, our understanding of the data used to train and evaluate these algorithms remains stagnant. In this work, we first identify and measure two distinct kinds of distribution shifts that are ubiquitous in various datasets. Next, through extensive experiments, we compare OoD generalization algorithms across two groups of benchmarks, each dominated by one of the distribution shifts, revealing their strengths on one shift as well as limitations on the other shift. Overall, we position existing datasets and algorithms from different research areas seemingly unconnected into the same coherent picture. It may serve as a foothold that can be resorted to by future OoD generalization research. Our code is available at https://github.com/ynysjtu/ood_bench.