Deep Subdomain Adaptation Network for Image Classification

TLDR

Domain adaptation transfers models to unlabeled target domains, but existing methods align only global distributions and ignore subdomain relationships, resulting in suboptimal performance; subdomain adaptation aims to align relevant subdomains but is often adversarial and slow. The authors propose DSAN, a network that aligns relevant subdomain distributions across domains using local maximum mean discrepancy (LMMD). DSAN extends feed‑forward models with an LMMD loss that aligns subdomain activations and can be trained efficiently by back‑propagation. DSAN is simple, non‑adversarial, fast to converge, and achieves remarkable results on object recognition and digit classification tasks. Code is available at https://github.com/easezyc/deep-transfer-learning.

Abstract

For a target task where labeled data is unavailable, domain adaptation can transfer a learner from a different source domain. Previous deep domain adaptation methods mainly learn a global domain shift, i.e., align the global source and target distributions without considering the relationships between two subdomains within the same category of different domains, leading to unsatisfying transfer learning performance without capturing the fine-grained information. Recently, more and more researchers pay attention to Subdomain Adaptation which focuses on accurately aligning the distributions of the relevant subdomains. However, most of them are adversarial methods which contain several loss functions and converge slowly. Based on this, we present Deep Subdomain Adaptation Network (DSAN) which learns a transfer network by aligning the relevant subdomain distributions of domain-specific layer activations across different domains based on a local maximum mean discrepancy (LMMD). Our DSAN is very simple but effective which does not need adversarial training and converges fast. The adaptation can be achieved easily with most feed-forward network models by extending them with LMMD loss, which can be trained efficiently via back-propagation. Experiments demonstrate that DSAN can achieve remarkable results on both object recognition tasks and digit classification tasks. Our code will be available at: https://github.com/easezyc/deep-transfer-learning