Meta Self-Refinement for Robust Learning with Weak Supervision
This addresses the challenge of reducing annotation costs in NLP by improving robustness to label noise, though it is incremental as it builds on existing self-training frameworks.
The paper tackles the problem of training deep neural networks with noisy labels from weak supervision, which can cause overfitting and poor generalization, by proposing Meta Self-Refinement (MSR) to refine pseudo-labels and reduce error propagation, resulting in up to 11.4% accuracy and 9.26% F1 score improvements over state-of-the-art methods on eight NLP benchmarks.
Training deep neural networks (DNNs) under weak supervision has attracted increasing research attention as it can significantly reduce the annotation cost. However, labels from weak supervision can be noisy, and the high capacity of DNNs enables them to easily overfit the label noise, resulting in poor generalization. Recent methods leverage self-training to build noise-resistant models, in which a teacher trained under weak supervision is used to provide highly confident labels for teaching the students. Nevertheless, the teacher derived from such frameworks may have fitted a substantial amount of noise and therefore produce incorrect pseudo-labels with high confidence, leading to severe error propagation. In this work, we propose Meta Self-Refinement (MSR), a noise-resistant learning framework, to effectively combat label noise from weak supervision. Instead of relying on a fixed teacher trained with noisy labels, we encourage the teacher to refine its pseudo-labels. At each training step, MSR performs a meta gradient descent on the current mini-batch to maximize the student performance on a clean validation set. Extensive experimentation on eight NLP benchmarks demonstrates that MSR is robust against label noise in all settings and outperforms state-of-the-art methods by up to 11.4% in accuracy and 9.26% in F1 score.