The existence of multiple datasets for sarcasm detection prompts us to apply transfer learning to exploit their commonality. The adversarial neural transfer (ANT) framework utilizes multiple loss terms that encourage the source-domain and the target-domain feature distributions to be similar while optimizing for domain-specific performance. However, these objectives may be in conflict, which can lead to optimization difficulties and sometimes diminished transfer. We propose a generalized latent optimization strategy that allows different losses to accommodate each other and improves training dynamics. The proposed method outperforms transfer learning and meta-learning baselines. In particular, we achieve 10.02% absolute performance gain over the previous state of the art on the iSarcasm dataset.

翻译：用于讽刺性探测的多个数据集的存在促使我们应用转移学习来利用它们的共性。对抗性神经转移(ANT)框架使用多个损失条件,鼓励源域和目标域特性分布相似,同时优化特定域的性能。然而,这些目标可能存在冲突,可能导致优化困难,有时甚至减少转移。我们提出了一个普遍潜伏优化战略,允许不同的损失相互适应,改善培训动态。拟议方法优于转移学习和元学习基线。特别是,我们比iSarcasm数据集的以往水平实现了1.02%的绝对性性性业绩增益。