Task-Adversarial Co-Generative Nets
Task-Adversarial Co-Generative Nets
复制标题
DOI：
10.1145/3292500.3330843
复制
发表时间：
2019-07
期刊：
影响因子：
--
通讯作者：
Pei Yang;Qi Tan;Hanghang Tong;Jingrui He
中科院分区：
文献类型：
--
作者：
Pei Yang;Qi Tan;Hanghang Tong;Jingrui He
在本文中，我们提出了任务-对抗协同生成网络（TAGN），用于从多个任务中学习。它旨在解决多任务学习的两个基本问题，即，域转移和有限的标记数据，以原则性的方式。为此，TAGN首先学习特征的任务不变表示，以桥接任务之间的域转移。基于任务不变特征，TAGN为每个任务生成合理的示例，以解决数据稀缺问题。在TAGN中，我们利用多个游戏玩家，通过使用对抗策略来逐步提高特征和示例的联合生成的质量。它同时学习跨不同任务的任务不变特征的边缘分布以及每个任务的标签示例的联合分布。理论研究表明，在多人博弈的均衡点上，特征提取器准确地为不同的任务产生任务不变特征，而生成器和分类器都完美地复制了每个任务的联合分布。在基准数据集上的实验结果证明了该方法的有效性。
In this paper, we propose Task-Adversarial co-Generative Nets (TAGN) for learning from multiple tasks. It aims to address the two fundamental issues of multi-task learning, i.e., domain shift and limited labeled data, in a principled way. To this end, TAGN first learns the task-invariant representations of features to bridge the domain shift among tasks. Based on the task-invariant features, TAGN generates the plausible examples for each task to tackle the data scarcity issue. In TAGN, we leverage multiple game players to gradually improve the quality of the co-generation of features and examples by using an adversarial strategy. It simultaneously learns the marginal distribution of task-invariant features across different tasks and the joint distributions of examples with labels for each task. The theoretical study shows the desired results: at the equilibrium point of the multi-player game, the feature extractor exactly produces the task-invariant features for different tasks, while both the generator and the classifier perfectly replicate the joint distribution for each task. The experimental results on the benchmark data sets demonstrate the effectiveness of the proposed approach.