Do-Operation Guided Causal Representation Learning with Reduced Supervision Strength

Autor:	Zhu, Jiageng, Xie, Hanchen, AbdAlmageed, Wael
Rok vydání:	2022
Předmět:	Computer Science - Machine Learning Computer Science - Artificial Intelligence Statistics - Methodology
Druh dokumentu:	Working Paper
Popis:	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. Comment: NeurIPS 2022 Workshop CML4Impact Workshop Camera Ready
Databáze:	arXiv
Externí odkaz:	http://arxiv.org/abs/2206.01802 Zobrazit plný text záznamu View this record from Arxiv