On $f$-Divergence Principled Domain Adaptation: An Improved Framework
This work provides incremental improvements to the theoretical foundations of domain adaptation, benefiting researchers and practitioners dealing with distribution shifts in machine learning.
The paper tackles the problem of distribution shifts in unsupervised domain adaptation by refining an existing f-divergence-based discrepancy and introducing a new measure, f-domain discrepancy (f-DD), which leads to improved theoretical bounds and superior empirical performance on popular benchmarks.
Unsupervised domain adaptation (UDA) plays a crucial role in addressing distribution shifts in machine learning. In this work, we improve the theoretical foundations of UDA proposed in Acuna et al. (2021) by refining their $f$-divergence-based discrepancy and additionally introducing a new measure, $f$-domain discrepancy ($f$-DD). By removing the absolute value function and incorporating a scaling parameter, $f$-DD obtains novel target error and sample complexity bounds, allowing us to recover previous KL-based results and bridging the gap between algorithms and theory presented in Acuna et al. (2021). Using a localization technique, we also develop a fast-rate generalization bound. Empirical results demonstrate the superior performance of $f$-DD-based learning algorithms over previous works in popular UDA benchmarks.