PhD student, Columbia University
1 paper at NeurIPS 2025
Statistical prediction may be sufficient to drive the emergence of internal causal models and causal inference capacities in deep neural networks.