Randomized Controlled Trial

Randomized controlled trials (RCTs) are the gold standard for establishing causal relationships, primarily aiming to accurately estimate treatment effects. Current research focuses on improving RCT efficiency and robustness through techniques like optimized data partitioning (e.g., using Wasserstein Homogeneity Partitioning), addressing interference effects in networked data (e.g., via Markov Decision Processes), and developing more efficient measurement strategies (e.g., using transformer-based planned missing designs). These advancements enhance the reliability and generalizability of RCT findings, impacting various fields from medicine and technology to social sciences by providing more accurate and cost-effective causal inference.

Papers