Proxy Methods for Domain Adaptation
This addresses domain adaptation challenges in machine learning for scenarios with distribution shifts due to latent confounders, representing an incremental improvement by applying existing causal techniques to new settings.
The paper tackles domain adaptation under distribution shift caused by unobserved latent confounders, using proxy variables and proximal causal learning to adapt without explicitly modeling latents, and demonstrates that their two-stage kernel estimation approach outperforms methods that recover latent confounders in experiments.
We study the problem of domain adaptation under distribution shift, where the shift is due to a change in the distribution of an unobserved, latent variable that confounds both the covariates and the labels. In this setting, neither the covariate shift nor the label shift assumptions apply. Our approach to adaptation employs proximal causal learning, a technique for estimating causal effects in settings where proxies of unobserved confounders are available. We demonstrate that proxy variables allow for adaptation to distribution shift without explicitly recovering or modeling latent variables. We consider two settings, (i) Concept Bottleneck: an additional ''concept'' variable is observed that mediates the relationship between the covariates and labels; (ii) Multi-domain: training data from multiple source domains is available, where each source domain exhibits a different distribution over the latent confounder. We develop a two-stage kernel estimation approach to adapt to complex distribution shifts in both settings. In our experiments, we show that our approach outperforms other methods, notably those which explicitly recover the latent confounder.