Domain Generalization via Causal Adjustment for Cross-Domain Sentiment Analysis
This addresses the problem of generalizing to unknown test domains in sentiment analysis, which is an incremental improvement over existing domain adaptation methods.
The paper tackles domain generalization for cross-domain sentiment analysis by proposing a backdoor adjustment-based causal model to disentangle domain-specific and domain-invariant representations, achieving great performance and robustness in experiments compared to state-of-the-art baselines.
Domain adaption has been widely adapted for cross-domain sentiment analysis to transfer knowledge from the source domain to the target domain. Whereas, most methods are proposed under the assumption that the target (test) domain is known, making them fail to generalize well on unknown test data that is not always available in practice. In this paper, we focus on the problem of domain generalization for cross-domain sentiment analysis. Specifically, we propose a backdoor adjustment-based causal model to disentangle the domain-specific and domain-invariant representations that play essential roles in tackling domain shift. First, we rethink the cross-domain sentiment analysis task in a causal view to model the causal-and-effect relationships among different variables. Then, to learn an invariant feature representation, we remove the effect of domain confounders (e.g., domain knowledge) using the backdoor adjustment. A series of experiments over many homologous and diverse datasets show the great performance and robustness of our model by comparing it with the state-of-the-art domain generalization baselines.