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 2206.01802cs.LGcs.AIstat.ME
收藏

减少监督强度的Do操作引导的因果表示学习

Do-Operation Guided Causal Representation Learning with Reduced Supervision Strength

USC(南加州大学);USC Information Sciences Institute(南加州大学信息科学研究所)

Jiageng Zhu, Hanchen Xie, Wael AbdAlmageed

摘要:

因果表示学习已经被提出来编码在高维数据中呈现的因素之间的关系。然而,现有的方法仅使用大量的标记数据,而忽略了由相同因果机制生成的样本遵循相同因果关系的事实。在本文中,我们试图探索这些信息,利用做操作,以减少监管力度。我们提出了一个框架,通过交换从一对输入编码的潜在原因和影响因素来实现做操作。此外,我们还确定了现有的因果表征指标的不足之处,经验和理论,并引入新的指标更好的评估。在人工数据集和真实数据集上进行的实验表明,与现有方法相比,该方法具有明显的优越性。

英文摘要:

Causal representation learning has been proposed to encode relationships between factors presented in the high dimensional data. However, existing methods suffer from merely using a large amount of labeled data and ignore the fact that samples generated by the same causal mechanism follow the same causal relationships. In this paper, we seek to explore such information by leveraging do-operation to reduce supervision strength. We propose a framework that implements do-operation by swapping latent cause and effect factors encoded from a pair of inputs. Moreover, we also identify the inadequacy of existing causal representation metrics empirically and theoretically and introduce new metrics for better evaluation. Experiments conducted on both synthetic and real datasets demonstrate the superiorities of our method compared with state-of-the-art methods.

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。

↑