Pre-training Text Representations as Meta Learning

by   Shangwen Lv, et al.

Pre-training text representations has recently been shown to significantly improve the state-of-the-art in many natural language processing tasks. The central goal of pre-training is to learn text representations that are useful for subsequent tasks. However, existing approaches are optimized by minimizing a proxy objective, such as the negative log likelihood of language modeling. In this work, we introduce a learning algorithm which directly optimizes model's ability to learn text representations for effective learning of downstream tasks. We show that there is an intrinsic connection between multi-task pre-training and model-agnostic meta-learning with a sequence of meta-train steps. The standard multi-task learning objective adopted in BERT is a special case of our learning algorithm where the depth of meta-train is zero. We study the problem in two settings: unsupervised pre-training and supervised pre-training with different pre-training objects to verify the generality of our approach.Experimental results show that our algorithm brings improvements and learns better initializations for a variety of downstream tasks.



There are no comments yet.


page 1

page 2

page 3

page 4


Learning a Better Initialization for Soft Prompts via Meta-Learning

Prompt tuning (PT) is an effective approach to adapting pre-trained lang...

Profile Prediction: An Alignment-Based Pre-Training Task for Protein Sequence Models

For protein sequence datasets, unlabeled data has greatly outpaced label...

Should We Be Pre-training? An Argument for End-task Aware Training as an Alternative

Pre-training, where models are trained on an auxiliary objective with ab...

On the Influence of Masking Policies in Intermediate Pre-training

Current NLP models are predominantly trained through a pretrain-then-fin...

MC-BERT: Efficient Language Pre-Training via a Meta Controller

Pre-trained contextual representations (e.g., BERT) have become the foun...

Learning Unsupervised Learning Rules

A major goal of unsupervised learning is to discover data representation...

Pre-Training Transformers as Energy-Based Cloze Models

We introduce Electric, an energy-based cloze model for representation le...
This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.