Our research in causal machine learning focuses on causal discovery and deep causal representation learning. In causal discovery, we develop methods to learn DAGs/Bayesian networks/SCMs from data, including both local and global methods, learning under insufficient data, and with heteroscedastic additive noise assumptions. For causal representation learning, we are developing methods by combining causal learning with deep learning through active intervention to learn deep causal representations that are robust to data selection bias, latent confounders, and generalize well across domains.