BERTs of a feather do not generalize together: Large variability in generalization across models with similar test set performance

TL;DR

This study reveals that multiple instances of the same neural network architecture trained on the same data can have highly variable generalization behaviors, despite similar performance on the training set.

Contribution

It demonstrates significant variability in generalization across models with similar test accuracy, highlighting the influence of local minima and the need for stronger inductive biases.

Findings

Models show consistent accuracy on MNLI development set.

Wide variability in syntactic generalization performance on HANS.

Variability likely due to local minima in training landscape.

Abstract

If the same neural network architecture is trained multiple times on the same dataset, will it make similar linguistic generalizations across runs? To study this question, we fine-tuned 100 instances of BERT on the Multi-genre Natural Language Inference (MNLI) dataset and evaluated them on the HANS dataset, which evaluates syntactic generalization in natural language inference. On the MNLI development set, the behavior of all instances was remarkably consistent, with accuracy ranging between 83.6% and 84.8%. In stark contrast, the same models varied widely in their generalization performance. For example, on the simple case of subject-object swap (e.g., determining that "the doctor visited the lawyer" does not entail "the lawyer visited the doctor"), accuracy ranged from 0.00% to 66.2%. Such variation is likely due to the presence of many local minima that are equally attractive to a low-bias learner such as a neural network; decreasing the variability may therefore require models with stronger inductive biases.