PyWhy Causality in Practice

PyWhy Causality in Practice is a new talk series focusing on causality and machine learning, especially from a practical perspective. We'll have tutorials and presentations about PyWhy libraries but also talks by external speakers working on causal inference.

Lessons learned from the DAGitty user community

Johannes Textor works both at the Radboud University and the Radboud University Medical Center in Nijmegen, The Netherlands. He is interested in leveraging causal inference methodology for the benefit of biomedical research, especially in the fields of Immunology and Tumor Immunology.

In this talk, Johannes describes his reflections on DAGitty usage by biomedical scientists and to what extent causal graphs are useful in science. In his words, "I started developing the tool “dagitty” in 2010, first as a website, and then as an R package. I don’t know exactly how many people use this tool, but I believe it’s a substantial amount: there are ~1000 visits to the site per day, ~17000 causal diagrams have been saved on the website so far, and the two dagitty papers have ~2800 citations. Over the years, feedback from the user base has provided me with unique insights into the users’ issues and priorities. More recently, I’ve also actively tried to get insight into how dagitty (and causal diagrams more broadly) are being used and if this is actually beneficial for science (I currently have my doubts). In the talk, I’lll share some stories of these interactions and how they shaped dagitty and myself over the years."

causal-learn library: Causal discovery in Python

Yujia Zheng, a Ph.D. student at CMU, talks about the causal-learn package and how it can be used to learn causal graphs (and more) from observational data.

Causal discovery aims at revealing causal relations from observational data, which is a fundamental task in science and engineering. This talk introduces causal-learn, an open-source Python library for causal discovery. This library focuses on bringing a comprehensive collection of causal discovery methods to both practitioners and researchers. It provides easy-to-use APIs for non-specialists, modular building blocks for developers, detailed documentation for learners, and comprehensive methods for all. Different from previous packages in R or Java, causal-learn is fully developed in Python, which could be more in tune with the recent preference shift in programming languages within related communities. The talk will also explore related usage examples, aiming to further lower the entry threshold by providing a roadmap for selecting the appropriate algorithm.

Causal Representation Learning: Discovery of the Hidden World

Prof. Kun Zhang, currently on leave from Carnegie Mellon University (CMU), is a professor and the acting chair of the machine learning department and the director of the Center for Integrative AI at Mohamed bin Zayed University of Artificial Intelligence (MBZUAI). In this talk, he gives an overview of causal representation learning and how it has evolved over time.

Causality is a fundamental notion in science, engineering, and even in machine learning. Causal representation learning aims to reveal the underlying high-level hidden causal variables and their relations. It can be seen as a special case of causal discovery, whose goal is to recover the underlying causal structure or causal model from observational data. The modularity property of a causal system implies properties of minimal changes and independent changes of causal representations, and in this talk, we show how such properties make it possible to recover the underlying causal representations from observational data with identifiability guarantees: under appropriate assumptions, the learned representations are consistent with the underlying causal process. Various problem settings are considered, involving independent and identically distributed (i.i.d.) data, temporal data, or data with distribution shift as input. We demonstrate when identifiable causal representation learning can benefit from flexible deep learning and when suitable parametric assumptions have to be imposed on the causal process, with various examples and applications.

EconML library and what's new in v0.15

EconML is a Python package that implements several cutting-edge causal inference estimators on top of flexible machine learning methods. In this talk, Keith Battocchi, software engineer at Microsoft Research New England and lead developer for EconML, presents a brief overview of EconML followed by a closer look at several big new features in EconML 0.15.

LLMs for causal inference

Emre Kiciman, Senior Principal Researcher at Microsoft, talks about pywhy-llm, a new experimental library that focuses on using large language models for causality.