HetMAML: Task-Heterogeneous Model-Agnostic Meta-Learning for Few-Shot Learning Across Modalities
HetMAML: Task-Heterogeneous Model-Agnostic Meta-Learning for Few-Shot Learning Across Modalities
复制标题
DOI：
10.1145/3459637.3482262
复制
发表时间：
2021-05
期刊：
影响因子：
--
通讯作者：
Jiayi Chen;Aidong Zhang
中科院分区：
文献类型：
--
作者：
Jiayi Chen;Aidong Zhang
大多数现有的基于梯度的元学习方法的少镜头学习假设，所有的任务有相同的输入特征空间。然而，在真实的世界场景中，存在任务的输入结构可以不同的许多情况，即，不同的任务可以在输入模态或数据类型的数量上变化。现有的元学习器不能处理异构任务分布（HTD），因为不仅有全局元知识共享的任务，但也类型特定的知识，区分每种类型的任务。为了处理任务的异构性和促进快速的任务内适应每种类型的任务，在本文中，我们提出了HetMAML，任务异构模型不可知的元学习框架，它可以捕获类型特定的和全局共享的知识，可以实现知识定制和泛化之间的平衡。具体来说，我们设计了一个多通道骨干模块，将每种类型的任务的输入编码成相同长度的特定于模态的嵌入序列。然后，我们提出了一个任务感知的迭代特征聚合网络，它可以自动考虑特定于任务的输入结构的上下文，并自适应地将异构输入空间投影到同一个低维的概念嵌入空间。我们在六个任务异构数据集上的实验表明，HetMAML成功地利用了异构任务的类型特定和全局共享的元参数，并为每种类型的任务实现了快速的任务内自适应。
Most of existing gradient-based meta-learning approaches to few-shot learning assume that all tasks have the same input feature space. However, in the real world scenarios, there are many cases that the input structures of tasks can be different, that is, different tasks may vary in the number of input modalities or data types. Existing meta-learners cannot handle the heterogeneous task distribution (HTD) as there is not only global meta-knowledge shared across tasks but also type-specific knowledge that distinguishes each type of tasks. To deal with task heterogeneity and promote fast within-task adaptions for each type of tasks, in this paper, we propose HetMAML, a task-heterogeneous model-agnostic meta-learning framework, which can capture both the type-specific and globally shared knowledge and can achieve the balance between knowledge customization and generalization. Specifically, we design a multi-channel backbone module that encodes the input of each type of tasks into the same length sequence of modality-specific embeddings. Then, we propose a task-aware iterative feature aggregation network which can automatically take into account the context of task-specific input structures and adaptively project the heterogeneous input spaces to the same lower-dimensional embedding space of concepts. Our experiments on six task-heterogeneous datasets demonstrate that HetMAML successfully leverages type-specific and globally shared meta-parameters for heterogeneous tasks and achieves fast within-task adaptions for each type of tasks.