RESUMEN
Domain adaptation aims to bridge the domain shifts between the source and the target domain. These shifts may span different dimensions such as fog, rainfall, etc. However, recent methods typically do not consider explicit prior knowledge about the domain shifts on a specific dimension, thus leading to less desired adaptation performance. In this article, we study a practical setting called Specific Domain Adaptation (SDA) that aligns the source and target domains in a demanded-specific dimension. Within this setting, we observe the intra-domain gap induced by different domainness (i.e., numerical magnitudes of domain shifts in this dimension) is crucial when adapting to a specific domain. To address the problem, we propose a novel Self-Adversarial Disentangling (SAD) framework. In particular, given a specific dimension, we first enrich the source domain by introducing a domainness creator with providing additional supervisory signals. Guided by the created domainness, we design a self-adversarial regularizer and two loss functions to jointly disentangle the latent representations into domainness-specific and domainness-invariant features, thus mitigating the intra-domain gap. Our method can be easily taken as a plug-and-play framework and does not introduce any extra costs in the inference time. We achieve consistent improvements over state-of-the-art methods in both object detection and semantic segmentation.
RESUMEN
Objective: To explore expression changes and clinical significance of EBI3 in gastric cancer. Methods: Expression of EBI3 in gastric cancer (GC) cell lines, GC tissues, and corresponding adjacent tissues were detected by qRT-PCR, Western blot, or immunohistochemistry. The relationship between the EBI3 expression and clinicopathological features of GC patients was analyzed. Expression of EBI3 in BGC-823 was overexpressed or downregulated, then, the changes of proliferation, migration, invasion, and tumorigenicity of BGC-823 were observed by MTT, scratch test, Transwell test, and tumorigenesis assay model. Results: EBI3 was lowly expressed in GC tissues. EBI3 expression in BGC823 was highest than other cell lines. EBI3 expression was significantly associated with TNM stage. GC patients with low expression of EBI3 had a rather poor prognosis than the GC patients with high expression of EBI3. Low EBI3 expression was an independent risk predictor of the prognosis of GC patients. After EBI3 was overexpressed, the viability, migration, invasion, and tumorigenicity abilities of BGC-823 were significantly reduced. Opposite effect was observed after EBI3 expression was downregulated. Conclusion: EBI3 low expression is closely related to the malignant degree of GC and may be a predictive indicator of the prognosis of GC and potential therapeutic targets.