Center for Targeted Machine Learning and Causal Inference

The Center for Targeted Machine Learning (CTML) is an interdisciplinary research center in UC Berkeley’s School of Public Health, combining cutting-edge causal inference, machine learning, and biostatistics to tackle pressing health challenges. As a beacon for rigorous, transparent, and reproducible science, CTML equips researchers with powerful tools to draw causal insights from randomized control trials (RCTs) and observational data—pivotal for evidence-based policy, clinical decisions, capacity building, and improving public health.

Key Research Initiatives & Partnerships

• Gilead–Berkeley Global Health Equity Initiative
  A landmark collaboration with Gilead Sciences to confront infectious and non‑communicable diseases worldwide. The initiative fuses applied research, biostatistics, robust data management, and executive education
   
• Joint Initiative for Causal Inference (with Novo Nordisk, University of Oxford, Harvard, University College London, University of Copenhagen, University of Ghent)
  A global alliance building methodological power for causal inference in RCTs and real-world data—positioning CTML at the forefront of international statistical best practices
   
• SEARCH Community Precision Health (via NIH/UCSF)
  The SEARCH consortium leverages CTML’s methods to drive multi-disease interventions aimed at eradicating AIDS and other diseases in East Africa.
   
• Sub-Saharan Data Science Grants (NIH support)
  Leveraging CTML’s faculty to build trauma, injury, and surgical equity research platforms across Africa

Educational Programs & Scientific Training

• Targeted Learning Courses & Seminars
  CTML offers advanced coursework (e.g. Targeted Learning, Survival Analysis, Causal Inference) taught by van der Laan, Hubbard,, Balzer, Schuler, Mertens and others—training students to deploy state-of-the-art estimators like TMLE and Super Learner
   
• Active Seminar & Reading Series
  The CTML Seminar Series engages faculty, students, and practitioners in weekly discussions—fostering collaboration and foregrounding breakthroughs such as deep LTMLE