arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2011.01418cs.LG
收藏

单目标域的元学习可迁移表示

Meta-learning Transferable Representations with a Single Target Domain

Stanford University(斯坦福大学)

Hong Liu, Jeff Z. HaoChen, Colin Wei, Tengyu Ma

摘要:

最近的研究发现,微调和联合训练-迁移学习的两种流行方法-并不总是提高下游任务的准确性。首先,我们的目标是更多地了解微调和联合训练何时以及为什么对迁移学习来说是次优的,甚至是有害的。我们设计了半合成数据集,其中源任务可以通过源特定功能或可转移功能来解决。我们观察到(1)预训练可能没有学习可转移特征的动机,(2)联合训练可能同时学习特定于源的特征并过拟合目标。其次,为了改善微调和联合训练,我们提出了Meta表示学习(MeRLin)来学习可转移的特征。MeRLin元学习表示,通过确保头部适合在具有目标训练数据的表示的顶部上也在目标验证数据上表现良好。我们还证明了MeRLin恢复目标地面实况模型与二次神经网络参数化和源分布,其中包含可转移和源特定的功能。在相同的分布上,预训练和联合训练可证明无法学习可转移的特征。MeRLin在各种真实世界视觉和NLP迁移学习基准测试中的表现优于以前最先进的迁移学习算法。

英文摘要:

Recent works found that fine-tuning and joint training---two popular approaches for transfer learning---do not always improve accuracy on downstream tasks. First, we aim to understand more about when and why fine-tuning and joint training can be suboptimal or even harmful for transfer learning. We design semi-synthetic datasets where the source task can be solved by either source-specific features or transferable features. We observe that (1) pre-training may not have incentive to learn transferable features and (2) joint training may simultaneously learn source-specific features and overfit to the target. Second, to improve over fine-tuning and joint training, we propose Meta Representation Learning (MeRLin) to learn transferable features. MeRLin meta-learns representations by ensuring that a head fit on top of the representations with target training data also performs well on target validation data. We also prove that MeRLin recovers the target ground-truth model with a quadratic neural net parameterization and a source distribution that contains both transferable and source-specific features. On the same distribution, pre-training and joint training provably fail to learn transferable features. MeRLin empirically outperforms previous state-of-the-art transfer learning algorithms on various real-world vision and NLP transfer learning benchmarks.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑