Don't Take the Easy Way Out: Ensemble Based Methods for Avoiding Known Dataset Biases
This addresses robustness issues for machine learning models in NLP and vision tasks, though it is incremental as it builds on prior bias-aware methods.
The paper tackles the problem of models relying on superficial dataset biases that fail to generalize out-of-domain, by proposing an ensemble-based method that trains a robust model to focus on more generalizable patterns, resulting in a 12-point gain on a visual question answering dataset and a 9-point gain on an adversarial test set.
State-of-the-art models often make use of superficial patterns in the data that do not generalize well to out-of-domain or adversarial settings. For example, textual entailment models often learn that particular key words imply entailment, irrespective of context, and visual question answering models learn to predict prototypical answers, without considering evidence in the image. In this paper, we show that if we have prior knowledge of such biases, we can train a model to be more robust to domain shift. Our method has two stages: we (1) train a naive model that makes predictions exclusively based on dataset biases, and (2) train a robust model as part of an ensemble with the naive one in order to encourage it to focus on other patterns in the data that are more likely to generalize. Experiments on five datasets with out-of-domain test sets show significantly improved robustness in all settings, including a 12 point gain on a changing priors visual question answering dataset and a 9 point gain on an adversarial question answering test set.