任意路数元学习
Any-Way Meta Learning
AI总结：
针对元学习受限于固定类别基数的问题，提出“any-way”学习范式，利用随机数字标签分配产生的标签等价性解放模型约束，并引入语义类别信息注入机制，在 MAML 和 ProtoNet 上验证了其优于传统固定类别基数模型的性能。
摘要：
尽管元学习在快速适应领域展现出极具前景的性能，但它受限于固定的类别基数。当面临训练期间未曾见过的、具有不同类别基数的任务时，模型缺乏相应的能力。在本文中，我们通过利用在情景任务采样期间随机数字标签分配所产生的“标签等价性”来解决并克服这一挑战。针对什么定义了“真正”的元学习的疑问，我们引入了“any-way”学习范式，这是一种创新的模型训练方法，将模型从固定类别基数约束中解放出来。令人惊讶的是，该模型在性能、收敛速度和稳定性方面不仅匹配，而且往往优于传统的固定类别基数模型。这颠覆了关于领域泛化的既有观念。此外，我们认为固有的标签等价性天然缺乏语义信息。为了弥合由标签等价性引发的这一语义信息鸿沟，我们进一步提出了一种将语义类别信息注入模型的机制。这将增强模型的理解力与功能性。在 MAML 和 ProtoNet 等知名架构上进行的实验证实了我们方法的有效性。
英文摘要：
Although meta-learning seems promising performance in the realm of rapid adaptability, it is constrained by fixed cardinality. When faced with tasks of varying cardinalities that were unseen during training, the model lacks its ability. In this paper, we address and resolve this challenge by harnessing `label equivalence' emerged from stochastic numeric label assignments during episodic task sampling. Questioning what defines ``true" meta-learning, we introduce the ``any-way" learning paradigm, an innovative model training approach that liberates model from fixed cardinality constraints. Surprisingly, this model not only matches but often outperforms traditional fixed-way models in terms of performance, convergence speed, and stability. This disrupts established notions about domain generalization. Furthermore, we argue that the inherent label equivalence naturally lacks semantic information. To bridge this semantic information gap arising from label equivalence, we further propose a mechanism for infusing semantic class information into the model. This would enhance the model's comprehension and functionality. Experiments conducted on renowned architectures like MAML and ProtoNet affirm the effectiveness of our method.
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。
