Weakly-Supervised Deep Learning for Domain Invariant Sentiment Classification

TL;DR

This paper proposes a two-stage weakly supervised training method for domain-invariant sentiment classification, enabling models to adapt to new domains without target domain data during training.

Contribution

It introduces a simple lift-and-shift approach using weak supervision, achieving near-target domain performance without target data exposure during training.

Findings

Performance close to supervised target domain training

Effective domain adaptation with weak supervision

No target domain data needed during training

Abstract

The task of learning a sentiment classification model that adapts well to any target domain, different from the source domain, is a challenging problem. Majority of the existing approaches focus on learning a common representation by leveraging both source and target data during training. In this paper, we introduce a two-stage training procedure that leverages weakly supervised datasets for developing simple lift-and-shift-based predictive models without being exposed to the target domain during the training phase. Experimental results show that transfer with weak supervision from a source domain to various target domains provides performance very close to that obtained via supervised training on the target domain itself.