Full Professor, Georgia Institute of Technology
1 paper at NeurIPS 2025
We propose DCCD-CONF, a scalable, theoretically grounded framework for learning nonlinear cyclic causal graphs with unmeasured confounders using interventional data.
